ChromHMM Tutorial - Genome.gov · ChromHMM Tutorial Jason Ernst Assistant Professor University of...
Transcript of ChromHMM Tutorial - Genome.gov · ChromHMM Tutorial Jason Ernst Assistant Professor University of...
ChromHMM Tutorial
Jason Ernst Assistant Professor
University of California Los Angeles
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Chroma=n Marks for Genome Annota=on100+ histone modifications Specificity in bull Histone protein bull Amino acid residue bull Chemical modification (eg
methyl acetylation) bull Number of occurrence of
the modifications
Examples H3K4me1 ndash Enhancers H3K4me3 ndash Promoters H3K27me3 ndash Repressive H3K9me3 ndash Repressive H3K36me3 ndash Transcribed
Histone Modifications can be Mapped Genome-wide with ChIP-seq
Image source httpnihroadmapnihgovepigenomics
From lsquochromaMn marksrsquo to lsquochromaMn statesrsquo
bull Learn de novosignificant13
Promoter states
Transcribed states combinaMonal andspaMal paOerns ofchromaMn marks
Active Intergenic
Repressed bull Reveal funcMonalelements evenwithout13 looking atsequence
bull Use13 for13 genome13 annotaMon
Ernst and Kellis Nat Biotech 2010
Our approach Mul=variate Hidden Markov Model(ChromHMM)13
Enhancer Gene Starts Gene -shy‐ Transcribed Region DNA
BinarizedchromaMnmarks Calledbased on apoissondistribuMon
Most13 likely HiddenState 2
H3K4me3 H3K36me3 H3K36me3 H3K36me3 H3K36me3 H3K4me1 H3K4me3 H3K4me1
H3K27ac H3K4me1
1 3 4 6 6 6 6 6 5 5 5
Unobserved
High Probability ChromaMn0713 Marks in State
0813 0813 200 base pair interval 1 4K27ac
3
2 0913
0913
0813 K4me1 H3K4me3
H3K4me3
H3K4me1 H3K4me1 All probabiliMesEmission distribution is a product of independent are learned from5Bernoulli random variables
0913 the data13
6H3K36me3
Binarization leads to explicit modeling of mark combinations and interpretable parameters 6
Ernst and Kellis Nat Biotech 2010 Ernst and Kellis Nature Methods 2012
ENCODE Study nine marks in nine human cell lines 9 marks 9 human13 cell13 types13 81 Chroma=n Mark13 TracksH3K4me1
H3K4me2
H3K4me3
H3K27ac
H3K9ac
H3K27me3
H4K20me1
H3K36me3
CTCF
+WCE
+RNA
HUVEC Umbilical vein endothelial
NHEK Keratinocytes
GM12878 Lymphoblastoid
K562 Myelogenous leukemia
HepG2 Liver carcinoma
NHLF Normal human lung fibroblast
HMEC Mammary epithelial cell
HSMM Skeletal muscle myoblasts
H1 Embryonic
x
Ernst et al Nature 2011
bull Learned jointly across cell types (virtual concatenation)
bull State definitions are common
bull State locations are dynamic
Brad Bernstein ENCODE Group HepG2 NHLF HMEC HSMM HUVEC NHEK GM12878 K562 H1
Chroma=n states13 dynamics13 across13 nine13 ENCODE cell types
bull Single annotation track for each cell type
bull Summarize cell-type activity at a glance bull Can study 9-cell activity pattern across Ernst et al Nature 2011
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Chroma=n States Defined Across 127 CellTissues Types
Roadmap Epigenomics Consortium et al Nature 2015 16 epigenomes from ENCODE 2
Chroma=n States13 Defined on Imputed DataMarks
ChromImpute methodErnst13 and Kellis13 Nature Biotech 2015
ChromHMMModels across Many Roadmap ENCODE13 Cell13 and Tissue Types13
127 CellTissue13 Types13
98 CellTissue13 Types13
H3K4me1 H3K4me1H3K4me3 H3K4me3H3K27me3 H3K27me3H3K9me3 H3K9me3H3K36me3 H3K36me3
H3K27ac
127 CellTissue13 Types13 H3K4me1 H3K9acH3K4me3 H4K20me1H3K27me3 H3K79me2H3K9me3 H3K4me2H3K36me3 H2AZH3K27ac DNase
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Note Different13 than track hubRoadmap Epigenomics Data13 Complete CollecMon at Wash U VizHub
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Chroma=n Marks for Genome Annota=on100+ histone modifications Specificity in bull Histone protein bull Amino acid residue bull Chemical modification (eg
methyl acetylation) bull Number of occurrence of
the modifications
Examples H3K4me1 ndash Enhancers H3K4me3 ndash Promoters H3K27me3 ndash Repressive H3K9me3 ndash Repressive H3K36me3 ndash Transcribed
Histone Modifications can be Mapped Genome-wide with ChIP-seq
Image source httpnihroadmapnihgovepigenomics
From lsquochromaMn marksrsquo to lsquochromaMn statesrsquo
bull Learn de novosignificant13
Promoter states
Transcribed states combinaMonal andspaMal paOerns ofchromaMn marks
Active Intergenic
Repressed bull Reveal funcMonalelements evenwithout13 looking atsequence
bull Use13 for13 genome13 annotaMon
Ernst and Kellis Nat Biotech 2010
Our approach Mul=variate Hidden Markov Model(ChromHMM)13
Enhancer Gene Starts Gene -shy‐ Transcribed Region DNA
BinarizedchromaMnmarks Calledbased on apoissondistribuMon
Most13 likely HiddenState 2
H3K4me3 H3K36me3 H3K36me3 H3K36me3 H3K36me3 H3K4me1 H3K4me3 H3K4me1
H3K27ac H3K4me1
1 3 4 6 6 6 6 6 5 5 5
Unobserved
High Probability ChromaMn0713 Marks in State
0813 0813 200 base pair interval 1 4K27ac
3
2 0913
0913
0813 K4me1 H3K4me3
H3K4me3
H3K4me1 H3K4me1 All probabiliMesEmission distribution is a product of independent are learned from5Bernoulli random variables
0913 the data13
6H3K36me3
Binarization leads to explicit modeling of mark combinations and interpretable parameters 6
Ernst and Kellis Nat Biotech 2010 Ernst and Kellis Nature Methods 2012
ENCODE Study nine marks in nine human cell lines 9 marks 9 human13 cell13 types13 81 Chroma=n Mark13 TracksH3K4me1
H3K4me2
H3K4me3
H3K27ac
H3K9ac
H3K27me3
H4K20me1
H3K36me3
CTCF
+WCE
+RNA
HUVEC Umbilical vein endothelial
NHEK Keratinocytes
GM12878 Lymphoblastoid
K562 Myelogenous leukemia
HepG2 Liver carcinoma
NHLF Normal human lung fibroblast
HMEC Mammary epithelial cell
HSMM Skeletal muscle myoblasts
H1 Embryonic
x
Ernst et al Nature 2011
bull Learned jointly across cell types (virtual concatenation)
bull State definitions are common
bull State locations are dynamic
Brad Bernstein ENCODE Group HepG2 NHLF HMEC HSMM HUVEC NHEK GM12878 K562 H1
Chroma=n states13 dynamics13 across13 nine13 ENCODE cell types
bull Single annotation track for each cell type
bull Summarize cell-type activity at a glance bull Can study 9-cell activity pattern across Ernst et al Nature 2011
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Chroma=n States Defined Across 127 CellTissues Types
Roadmap Epigenomics Consortium et al Nature 2015 16 epigenomes from ENCODE 2
Chroma=n States13 Defined on Imputed DataMarks
ChromImpute methodErnst13 and Kellis13 Nature Biotech 2015
ChromHMMModels across Many Roadmap ENCODE13 Cell13 and Tissue Types13
127 CellTissue13 Types13
98 CellTissue13 Types13
H3K4me1 H3K4me1H3K4me3 H3K4me3H3K27me3 H3K27me3H3K9me3 H3K9me3H3K36me3 H3K36me3
H3K27ac
127 CellTissue13 Types13 H3K4me1 H3K9acH3K4me3 H4K20me1H3K27me3 H3K79me2H3K9me3 H3K4me2H3K36me3 H2AZH3K27ac DNase
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Note Different13 than track hubRoadmap Epigenomics Data13 Complete CollecMon at Wash U VizHub
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Chroma=n Marks for Genome Annota=on100+ histone modifications Specificity in bull Histone protein bull Amino acid residue bull Chemical modification (eg
methyl acetylation) bull Number of occurrence of
the modifications
Examples H3K4me1 ndash Enhancers H3K4me3 ndash Promoters H3K27me3 ndash Repressive H3K9me3 ndash Repressive H3K36me3 ndash Transcribed
Histone Modifications can be Mapped Genome-wide with ChIP-seq
Image source httpnihroadmapnihgovepigenomics
From lsquochromaMn marksrsquo to lsquochromaMn statesrsquo
bull Learn de novosignificant13
Promoter states
Transcribed states combinaMonal andspaMal paOerns ofchromaMn marks
Active Intergenic
Repressed bull Reveal funcMonalelements evenwithout13 looking atsequence
bull Use13 for13 genome13 annotaMon
Ernst and Kellis Nat Biotech 2010
Our approach Mul=variate Hidden Markov Model(ChromHMM)13
Enhancer Gene Starts Gene -shy‐ Transcribed Region DNA
BinarizedchromaMnmarks Calledbased on apoissondistribuMon
Most13 likely HiddenState 2
H3K4me3 H3K36me3 H3K36me3 H3K36me3 H3K36me3 H3K4me1 H3K4me3 H3K4me1
H3K27ac H3K4me1
1 3 4 6 6 6 6 6 5 5 5
Unobserved
High Probability ChromaMn0713 Marks in State
0813 0813 200 base pair interval 1 4K27ac
3
2 0913
0913
0813 K4me1 H3K4me3
H3K4me3
H3K4me1 H3K4me1 All probabiliMesEmission distribution is a product of independent are learned from5Bernoulli random variables
0913 the data13
6H3K36me3
Binarization leads to explicit modeling of mark combinations and interpretable parameters 6
Ernst and Kellis Nat Biotech 2010 Ernst and Kellis Nature Methods 2012
ENCODE Study nine marks in nine human cell lines 9 marks 9 human13 cell13 types13 81 Chroma=n Mark13 TracksH3K4me1
H3K4me2
H3K4me3
H3K27ac
H3K9ac
H3K27me3
H4K20me1
H3K36me3
CTCF
+WCE
+RNA
HUVEC Umbilical vein endothelial
NHEK Keratinocytes
GM12878 Lymphoblastoid
K562 Myelogenous leukemia
HepG2 Liver carcinoma
NHLF Normal human lung fibroblast
HMEC Mammary epithelial cell
HSMM Skeletal muscle myoblasts
H1 Embryonic
x
Ernst et al Nature 2011
bull Learned jointly across cell types (virtual concatenation)
bull State definitions are common
bull State locations are dynamic
Brad Bernstein ENCODE Group HepG2 NHLF HMEC HSMM HUVEC NHEK GM12878 K562 H1
Chroma=n states13 dynamics13 across13 nine13 ENCODE cell types
bull Single annotation track for each cell type
bull Summarize cell-type activity at a glance bull Can study 9-cell activity pattern across Ernst et al Nature 2011
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Chroma=n States Defined Across 127 CellTissues Types
Roadmap Epigenomics Consortium et al Nature 2015 16 epigenomes from ENCODE 2
Chroma=n States13 Defined on Imputed DataMarks
ChromImpute methodErnst13 and Kellis13 Nature Biotech 2015
ChromHMMModels across Many Roadmap ENCODE13 Cell13 and Tissue Types13
127 CellTissue13 Types13
98 CellTissue13 Types13
H3K4me1 H3K4me1H3K4me3 H3K4me3H3K27me3 H3K27me3H3K9me3 H3K9me3H3K36me3 H3K36me3
H3K27ac
127 CellTissue13 Types13 H3K4me1 H3K9acH3K4me3 H4K20me1H3K27me3 H3K79me2H3K9me3 H3K4me2H3K36me3 H2AZH3K27ac DNase
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Note Different13 than track hubRoadmap Epigenomics Data13 Complete CollecMon at Wash U VizHub
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Chroma=n Marks for Genome Annota=on100+ histone modifications Specificity in bull Histone protein bull Amino acid residue bull Chemical modification (eg
methyl acetylation) bull Number of occurrence of
the modifications
Examples H3K4me1 ndash Enhancers H3K4me3 ndash Promoters H3K27me3 ndash Repressive H3K9me3 ndash Repressive H3K36me3 ndash Transcribed
Histone Modifications can be Mapped Genome-wide with ChIP-seq
Image source httpnihroadmapnihgovepigenomics
From lsquochromaMn marksrsquo to lsquochromaMn statesrsquo
bull Learn de novosignificant13
Promoter states
Transcribed states combinaMonal andspaMal paOerns ofchromaMn marks
Active Intergenic
Repressed bull Reveal funcMonalelements evenwithout13 looking atsequence
bull Use13 for13 genome13 annotaMon
Ernst and Kellis Nat Biotech 2010
Our approach Mul=variate Hidden Markov Model(ChromHMM)13
Enhancer Gene Starts Gene -shy‐ Transcribed Region DNA
BinarizedchromaMnmarks Calledbased on apoissondistribuMon
Most13 likely HiddenState 2
H3K4me3 H3K36me3 H3K36me3 H3K36me3 H3K36me3 H3K4me1 H3K4me3 H3K4me1
H3K27ac H3K4me1
1 3 4 6 6 6 6 6 5 5 5
Unobserved
High Probability ChromaMn0713 Marks in State
0813 0813 200 base pair interval 1 4K27ac
3
2 0913
0913
0813 K4me1 H3K4me3
H3K4me3
H3K4me1 H3K4me1 All probabiliMesEmission distribution is a product of independent are learned from5Bernoulli random variables
0913 the data13
6H3K36me3
Binarization leads to explicit modeling of mark combinations and interpretable parameters 6
Ernst and Kellis Nat Biotech 2010 Ernst and Kellis Nature Methods 2012
ENCODE Study nine marks in nine human cell lines 9 marks 9 human13 cell13 types13 81 Chroma=n Mark13 TracksH3K4me1
H3K4me2
H3K4me3
H3K27ac
H3K9ac
H3K27me3
H4K20me1
H3K36me3
CTCF
+WCE
+RNA
HUVEC Umbilical vein endothelial
NHEK Keratinocytes
GM12878 Lymphoblastoid
K562 Myelogenous leukemia
HepG2 Liver carcinoma
NHLF Normal human lung fibroblast
HMEC Mammary epithelial cell
HSMM Skeletal muscle myoblasts
H1 Embryonic
x
Ernst et al Nature 2011
bull Learned jointly across cell types (virtual concatenation)
bull State definitions are common
bull State locations are dynamic
Brad Bernstein ENCODE Group HepG2 NHLF HMEC HSMM HUVEC NHEK GM12878 K562 H1
Chroma=n states13 dynamics13 across13 nine13 ENCODE cell types
bull Single annotation track for each cell type
bull Summarize cell-type activity at a glance bull Can study 9-cell activity pattern across Ernst et al Nature 2011
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Chroma=n States Defined Across 127 CellTissues Types
Roadmap Epigenomics Consortium et al Nature 2015 16 epigenomes from ENCODE 2
Chroma=n States13 Defined on Imputed DataMarks
ChromImpute methodErnst13 and Kellis13 Nature Biotech 2015
ChromHMMModels across Many Roadmap ENCODE13 Cell13 and Tissue Types13
127 CellTissue13 Types13
98 CellTissue13 Types13
H3K4me1 H3K4me1H3K4me3 H3K4me3H3K27me3 H3K27me3H3K9me3 H3K9me3H3K36me3 H3K36me3
H3K27ac
127 CellTissue13 Types13 H3K4me1 H3K9acH3K4me3 H4K20me1H3K27me3 H3K79me2H3K9me3 H3K4me2H3K36me3 H2AZH3K27ac DNase
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Note Different13 than track hubRoadmap Epigenomics Data13 Complete CollecMon at Wash U VizHub
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
From lsquochromaMn marksrsquo to lsquochromaMn statesrsquo
bull Learn de novosignificant13
Promoter states
Transcribed states combinaMonal andspaMal paOerns ofchromaMn marks
Active Intergenic
Repressed bull Reveal funcMonalelements evenwithout13 looking atsequence
bull Use13 for13 genome13 annotaMon
Ernst and Kellis Nat Biotech 2010
Our approach Mul=variate Hidden Markov Model(ChromHMM)13
Enhancer Gene Starts Gene -shy‐ Transcribed Region DNA
BinarizedchromaMnmarks Calledbased on apoissondistribuMon
Most13 likely HiddenState 2
H3K4me3 H3K36me3 H3K36me3 H3K36me3 H3K36me3 H3K4me1 H3K4me3 H3K4me1
H3K27ac H3K4me1
1 3 4 6 6 6 6 6 5 5 5
Unobserved
High Probability ChromaMn0713 Marks in State
0813 0813 200 base pair interval 1 4K27ac
3
2 0913
0913
0813 K4me1 H3K4me3
H3K4me3
H3K4me1 H3K4me1 All probabiliMesEmission distribution is a product of independent are learned from5Bernoulli random variables
0913 the data13
6H3K36me3
Binarization leads to explicit modeling of mark combinations and interpretable parameters 6
Ernst and Kellis Nat Biotech 2010 Ernst and Kellis Nature Methods 2012
ENCODE Study nine marks in nine human cell lines 9 marks 9 human13 cell13 types13 81 Chroma=n Mark13 TracksH3K4me1
H3K4me2
H3K4me3
H3K27ac
H3K9ac
H3K27me3
H4K20me1
H3K36me3
CTCF
+WCE
+RNA
HUVEC Umbilical vein endothelial
NHEK Keratinocytes
GM12878 Lymphoblastoid
K562 Myelogenous leukemia
HepG2 Liver carcinoma
NHLF Normal human lung fibroblast
HMEC Mammary epithelial cell
HSMM Skeletal muscle myoblasts
H1 Embryonic
x
Ernst et al Nature 2011
bull Learned jointly across cell types (virtual concatenation)
bull State definitions are common
bull State locations are dynamic
Brad Bernstein ENCODE Group HepG2 NHLF HMEC HSMM HUVEC NHEK GM12878 K562 H1
Chroma=n states13 dynamics13 across13 nine13 ENCODE cell types
bull Single annotation track for each cell type
bull Summarize cell-type activity at a glance bull Can study 9-cell activity pattern across Ernst et al Nature 2011
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Chroma=n States Defined Across 127 CellTissues Types
Roadmap Epigenomics Consortium et al Nature 2015 16 epigenomes from ENCODE 2
Chroma=n States13 Defined on Imputed DataMarks
ChromImpute methodErnst13 and Kellis13 Nature Biotech 2015
ChromHMMModels across Many Roadmap ENCODE13 Cell13 and Tissue Types13
127 CellTissue13 Types13
98 CellTissue13 Types13
H3K4me1 H3K4me1H3K4me3 H3K4me3H3K27me3 H3K27me3H3K9me3 H3K9me3H3K36me3 H3K36me3
H3K27ac
127 CellTissue13 Types13 H3K4me1 H3K9acH3K4me3 H4K20me1H3K27me3 H3K79me2H3K9me3 H3K4me2H3K36me3 H2AZH3K27ac DNase
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Note Different13 than track hubRoadmap Epigenomics Data13 Complete CollecMon at Wash U VizHub
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Our approach Mul=variate Hidden Markov Model(ChromHMM)13
Enhancer Gene Starts Gene -shy‐ Transcribed Region DNA
BinarizedchromaMnmarks Calledbased on apoissondistribuMon
Most13 likely HiddenState 2
H3K4me3 H3K36me3 H3K36me3 H3K36me3 H3K36me3 H3K4me1 H3K4me3 H3K4me1
H3K27ac H3K4me1
1 3 4 6 6 6 6 6 5 5 5
Unobserved
High Probability ChromaMn0713 Marks in State
0813 0813 200 base pair interval 1 4K27ac
3
2 0913
0913
0813 K4me1 H3K4me3
H3K4me3
H3K4me1 H3K4me1 All probabiliMesEmission distribution is a product of independent are learned from5Bernoulli random variables
0913 the data13
6H3K36me3
Binarization leads to explicit modeling of mark combinations and interpretable parameters 6
Ernst and Kellis Nat Biotech 2010 Ernst and Kellis Nature Methods 2012
ENCODE Study nine marks in nine human cell lines 9 marks 9 human13 cell13 types13 81 Chroma=n Mark13 TracksH3K4me1
H3K4me2
H3K4me3
H3K27ac
H3K9ac
H3K27me3
H4K20me1
H3K36me3
CTCF
+WCE
+RNA
HUVEC Umbilical vein endothelial
NHEK Keratinocytes
GM12878 Lymphoblastoid
K562 Myelogenous leukemia
HepG2 Liver carcinoma
NHLF Normal human lung fibroblast
HMEC Mammary epithelial cell
HSMM Skeletal muscle myoblasts
H1 Embryonic
x
Ernst et al Nature 2011
bull Learned jointly across cell types (virtual concatenation)
bull State definitions are common
bull State locations are dynamic
Brad Bernstein ENCODE Group HepG2 NHLF HMEC HSMM HUVEC NHEK GM12878 K562 H1
Chroma=n states13 dynamics13 across13 nine13 ENCODE cell types
bull Single annotation track for each cell type
bull Summarize cell-type activity at a glance bull Can study 9-cell activity pattern across Ernst et al Nature 2011
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Chroma=n States Defined Across 127 CellTissues Types
Roadmap Epigenomics Consortium et al Nature 2015 16 epigenomes from ENCODE 2
Chroma=n States13 Defined on Imputed DataMarks
ChromImpute methodErnst13 and Kellis13 Nature Biotech 2015
ChromHMMModels across Many Roadmap ENCODE13 Cell13 and Tissue Types13
127 CellTissue13 Types13
98 CellTissue13 Types13
H3K4me1 H3K4me1H3K4me3 H3K4me3H3K27me3 H3K27me3H3K9me3 H3K9me3H3K36me3 H3K36me3
H3K27ac
127 CellTissue13 Types13 H3K4me1 H3K9acH3K4me3 H4K20me1H3K27me3 H3K79me2H3K9me3 H3K4me2H3K36me3 H2AZH3K27ac DNase
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Note Different13 than track hubRoadmap Epigenomics Data13 Complete CollecMon at Wash U VizHub
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
ENCODE Study nine marks in nine human cell lines 9 marks 9 human13 cell13 types13 81 Chroma=n Mark13 TracksH3K4me1
H3K4me2
H3K4me3
H3K27ac
H3K9ac
H3K27me3
H4K20me1
H3K36me3
CTCF
+WCE
+RNA
HUVEC Umbilical vein endothelial
NHEK Keratinocytes
GM12878 Lymphoblastoid
K562 Myelogenous leukemia
HepG2 Liver carcinoma
NHLF Normal human lung fibroblast
HMEC Mammary epithelial cell
HSMM Skeletal muscle myoblasts
H1 Embryonic
x
Ernst et al Nature 2011
bull Learned jointly across cell types (virtual concatenation)
bull State definitions are common
bull State locations are dynamic
Brad Bernstein ENCODE Group HepG2 NHLF HMEC HSMM HUVEC NHEK GM12878 K562 H1
Chroma=n states13 dynamics13 across13 nine13 ENCODE cell types
bull Single annotation track for each cell type
bull Summarize cell-type activity at a glance bull Can study 9-cell activity pattern across Ernst et al Nature 2011
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Chroma=n States Defined Across 127 CellTissues Types
Roadmap Epigenomics Consortium et al Nature 2015 16 epigenomes from ENCODE 2
Chroma=n States13 Defined on Imputed DataMarks
ChromImpute methodErnst13 and Kellis13 Nature Biotech 2015
ChromHMMModels across Many Roadmap ENCODE13 Cell13 and Tissue Types13
127 CellTissue13 Types13
98 CellTissue13 Types13
H3K4me1 H3K4me1H3K4me3 H3K4me3H3K27me3 H3K27me3H3K9me3 H3K9me3H3K36me3 H3K36me3
H3K27ac
127 CellTissue13 Types13 H3K4me1 H3K9acH3K4me3 H4K20me1H3K27me3 H3K79me2H3K9me3 H3K4me2H3K36me3 H2AZH3K27ac DNase
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Note Different13 than track hubRoadmap Epigenomics Data13 Complete CollecMon at Wash U VizHub
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Chroma=n states13 dynamics13 across13 nine13 ENCODE cell types
bull Single annotation track for each cell type
bull Summarize cell-type activity at a glance bull Can study 9-cell activity pattern across Ernst et al Nature 2011
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Chroma=n States Defined Across 127 CellTissues Types
Roadmap Epigenomics Consortium et al Nature 2015 16 epigenomes from ENCODE 2
Chroma=n States13 Defined on Imputed DataMarks
ChromImpute methodErnst13 and Kellis13 Nature Biotech 2015
ChromHMMModels across Many Roadmap ENCODE13 Cell13 and Tissue Types13
127 CellTissue13 Types13
98 CellTissue13 Types13
H3K4me1 H3K4me1H3K4me3 H3K4me3H3K27me3 H3K27me3H3K9me3 H3K9me3H3K36me3 H3K36me3
H3K27ac
127 CellTissue13 Types13 H3K4me1 H3K9acH3K4me3 H4K20me1H3K27me3 H3K79me2H3K9me3 H3K4me2H3K36me3 H2AZH3K27ac DNase
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Note Different13 than track hubRoadmap Epigenomics Data13 Complete CollecMon at Wash U VizHub
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Chroma=n States Defined Across 127 CellTissues Types
Roadmap Epigenomics Consortium et al Nature 2015 16 epigenomes from ENCODE 2
Chroma=n States13 Defined on Imputed DataMarks
ChromImpute methodErnst13 and Kellis13 Nature Biotech 2015
ChromHMMModels across Many Roadmap ENCODE13 Cell13 and Tissue Types13
127 CellTissue13 Types13
98 CellTissue13 Types13
H3K4me1 H3K4me1H3K4me3 H3K4me3H3K27me3 H3K27me3H3K9me3 H3K9me3H3K36me3 H3K36me3
H3K27ac
127 CellTissue13 Types13 H3K4me1 H3K9acH3K4me3 H4K20me1H3K27me3 H3K79me2H3K9me3 H3K4me2H3K36me3 H2AZH3K27ac DNase
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Note Different13 than track hubRoadmap Epigenomics Data13 Complete CollecMon at Wash U VizHub
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Chroma=n States Defined Across 127 CellTissues Types
Roadmap Epigenomics Consortium et al Nature 2015 16 epigenomes from ENCODE 2
Chroma=n States13 Defined on Imputed DataMarks
ChromImpute methodErnst13 and Kellis13 Nature Biotech 2015
ChromHMMModels across Many Roadmap ENCODE13 Cell13 and Tissue Types13
127 CellTissue13 Types13
98 CellTissue13 Types13
H3K4me1 H3K4me1H3K4me3 H3K4me3H3K27me3 H3K27me3H3K9me3 H3K9me3H3K36me3 H3K36me3
H3K27ac
127 CellTissue13 Types13 H3K4me1 H3K9acH3K4me3 H4K20me1H3K27me3 H3K79me2H3K9me3 H3K4me2H3K36me3 H2AZH3K27ac DNase
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Note Different13 than track hubRoadmap Epigenomics Data13 Complete CollecMon at Wash U VizHub
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Chroma=n States13 Defined on Imputed DataMarks
ChromImpute methodErnst13 and Kellis13 Nature Biotech 2015
ChromHMMModels across Many Roadmap ENCODE13 Cell13 and Tissue Types13
127 CellTissue13 Types13
98 CellTissue13 Types13
H3K4me1 H3K4me1H3K4me3 H3K4me3H3K27me3 H3K27me3H3K9me3 H3K9me3H3K36me3 H3K36me3
H3K27ac
127 CellTissue13 Types13 H3K4me1 H3K9acH3K4me3 H4K20me1H3K27me3 H3K79me2H3K9me3 H3K4me2H3K36me3 H2AZH3K27ac DNase
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Note Different13 than track hubRoadmap Epigenomics Data13 Complete CollecMon at Wash U VizHub
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
ChromHMMModels across Many Roadmap ENCODE13 Cell13 and Tissue Types13
127 CellTissue13 Types13
98 CellTissue13 Types13
H3K4me1 H3K4me1H3K4me3 H3K4me3H3K27me3 H3K27me3H3K9me3 H3K9me3H3K36me3 H3K36me3
H3K27ac
127 CellTissue13 Types13 H3K4me1 H3K9acH3K4me3 H4K20me1H3K27me3 H3K79me2H3K9me3 H3K4me2H3K36me3 H2AZH3K27ac DNase
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Note Different13 than track hubRoadmap Epigenomics Data13 Complete CollecMon at Wash U VizHub
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Note Different13 than track hubRoadmap Epigenomics Data13 Complete CollecMon at Wash U VizHub
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Roadmap Epigenomics IntegraMve Analysis Portal
hOpcompbiomiteduroadmap
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Note Different13 than track hubRoadmap Epigenomics Data13 Complete CollecMon at Wash U VizHub
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Note Different13 than track hubRoadmap Epigenomics Data13 Complete CollecMon at Wash U VizHub
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
hOpgenomeucscedu13
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Note Different13 than track hubRoadmap Epigenomics Data13 Complete CollecMon at Wash U VizHub
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Note Different13 than track hubRoadmap Epigenomics Data13 Complete CollecMon at Wash U VizHub
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Accessing Roadmap ChromHMM throughthe UCSC Genome Browser Track Hubs
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Human Epigenome Browser atWashington University
hOpepigenomegatewaywustledu
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Talk Outline
bull ChromaMn states analysis and ChromHMM13 bull Accessing chromaMn state annotaMons forENCODE2 and Roadmap Epigenomics
bull Running the ChromHMM soware
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Soware download
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Soware manual
ChromHMMWebsitehOpcompbiomiteduChromHMM13
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Try to Run ChromHMM on SampleData13 on Your Computer
(Java13 needs to already be installed)1 DownloadhOpcompbiomiteduChromHMMChromHMMzip2 Unzip ChromHMMzip3 Open a command line4 Change into the ChromHMM directory5 Enter the commandjava -mx1600M -jar ChromHMMjar LearnModel -p 0 SAMPLEDATA_HG18 OUTPUTSAMPLE 10 hg18
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Input13 to ChromHMM
bull ChromHMMmodels are learned frombinarized data13 using its LearnModel command
bull Binarized data13 is typically obtained starMngfrom aligned readsndash Apply BinarizeBed if reads are in BED format13 ndash Apply BinarizeBam if reads are in BAM13 format
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
BinarizeBed
Java13 command lsquo-shy‐mx1600Mrsquo specifies memory to Java13
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
BinarizeBed
ChromHMM command
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
BinarizeBed
File with the chromosome lengths for the assembly
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
BinarizeBed
DIRECTORY of BED files
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
BinarizeBed
Cell-shy‐mark ndashfile table
cell mark cell-shy‐mark
Control data13 ndash is opMonal and canalso be treated as a mark
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
BinarizeBed
Output13 directory
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
LearnModel
lsquo-shy‐p 0rsquo Use as many processors as availablelsquo-shy‐p Nrsquo Use up to N processors (default13 N=1)
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
LearnModel
Directory with the Binarized Input13
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
LearnModel
Directory where the output13 goes
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
LearnModel
Number of states
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
LearnModel
Genome assembly
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
ChromHMM Report13
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
ChromHMM Report13
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Emission Parameters
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
TransiMon Parameters
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Model Parameter File
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
SegmentaMon File
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Browser Files13
hOpswwwbroadinsMtuteorgigv
Can load into browser UCSC Genome IGV
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Enrichments
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
PosiMonal Plots
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Enrichments for AddiMonal Cell Types
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Chroma=n states13 to interpret disease13 variants
bull Specific chromatin states enriched in GWAS catalog
Ernst and Kellis Nature Biotech 2010
bull Imputation based chromatin state used in dissection FTO loci
Claussnitzer et al NEJM 2015
bull Enhancers from different cell types enriched in different traits
Ernst et al Nature 2011
bull Interpreting epigenetic disease associated variation in Alzheimerrsquos disease
De Jager et al Nature Neuroscience 2014
bull Many other examples in the literature
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
Collaborators and Acknowledgements
bull Manolis Kellis
ENCODE consortium ndash Brad Bernstein production group
Roadmap Epigenomics consortium
Funding bull NHGRI NIH NSF HHMI Sloan Foundation
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
AddiMonal Commands
bull CompareModels ndash the command allows thecomparison of the emission parameters of a selectedmodel to a set13 of models in terms of correlaMon
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
AddiMonal Commands
bull MakeBrowserFiles ndash (re)generates browserfiles from segmentaMon files and allowsspecifying the coloring
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
AddiMonal Commands
bull OverlapEnrichment ndash (re)computesenrichments of a segmentaMon for a set13 ofannotaMons
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
AddiMonal Commands
bull NeighborhoodEnrichment ndash (re)computesenrichments of a segmentaMon around a set13 of anchor posiMons
AddiMonal Commands
bull Reorder ndash reorders the states of the model
AddiMonal Commands
bull Reorder ndash reorders the states of the model