PharmaMatrixWorkshop2010 Bioinforma6cDatabasespwinter/Bioinformatic...manual#and#automa=c#...

PharmaMatrix Workshop 2010 Bioinforma6c Databases

14 July 2010 Philip Winter & Ishwar Hosamani

Database Growth

Source: http://www.kokocinski.net/bioinformatics/databases.php

Genes & Proteins

Gene & Protein Interac=ons

Cheminforma=cs: Drugs & Metabolites

Database Survey

Genes & Proteins



UniProt

GenBank

dbSNP

PDB

GEO

Pfam TGI

Database Survey

Genes & Proteins



ZINC

UniProt

GenBank

DrugBank

dbSNP

PDB

GEO

PubChem

SciFinder

Pfam TGI

Database Survey

Database Survey

Genes & Proteins



ZINC

UniProt

GenBank

DrugBank

dbSNP

PDB

GEO

PubChem

SciFinder

BioGRID

Pfam TGI

KEGG

NetPath

Cura6on

•  Manual cura6on (or just cura6on): A human creates and annotates the database entry

•  Automa6c cura6on: A computer program creates and annotates the database entry

•  Semi-‐automa6c cura6on: A combina=on of manual and automa=c

Database Iden6fiers

•  Every database record will have a unique iden6fier; oUen this will be called an accession number which is assigned with the record is first added to the database

•  Be careful: databases will oUen permit a record to be modified but keep the same accession number; you should record the version number as well

•  Furthermore, databases may have different rules for handling records that are merged or split

Database Iden6fier Cheat Sheets

PaMern Iden6fier Name

Examples En6ty Database URL

[op=onal “GI:”][digits]

GenInfo Iden=fier

GI:34222261

Nucleo=de or protein sequence

GenBank, RefSeq

h`p://www.ncbi.nlm.nih.gov/

[leèr][5 digits] OR [2 leèrs][6 digits]

GenBank ACCESSION

AB088100 Nucleo=de sequence

GenBank h`p://www.ncbi.nlm.nih.gov/

[2 leèr type code]_[digits]

RefSeq ACCESSION

NM_178014 Nucleo=de or protein sequence

RefSeq h`p://www.ncbi.nlm.nih.gov/

[GenBank or RefSeq ACCESSION].[version number]

GenBank or RefSeq VERSION

AB088100.1

NM_178014.2


GenBank, RefSeq


(iden=cal to accession for recent entries)

GenBank LOCUS


GenBank, RefSeq




[Protein code]_[Species code]

Swiss-‐Prot ID (entry name)

TBB5_ HUMAN

Protein sequence

UniProtKB/Swiss-‐Prot

h`p://www.uniprot.org/

[UniProt AC]_[Species code]

UniProt ID (entry name)

Q9BUU9_ HUMAN

Protein sequence

UniProtKB/TrEMBL

h`p://www.uniprot.org/

[A-‐N,R-‐Z][0-‐9][A-‐Z][A-‐Z, 0-‐9][A-‐Z, 0-‐9][0-‐9] OR [O,P,Q][0-‐9][A-‐Z, 0-‐9][A-‐Z, 0-‐9][A-‐Z, 0-‐9][0-‐9]

UniProt AC (accession number)

P07437 Protein sequence

UniProtKB h`p://www.uniprot.org/



[capital le`ers or digits; no ini=al digit]

HGNC gene symbol

TUBB

TUBB1

Human gene

HGNC database

h`p://www.genenames.org/

GO:[7 digits]

GO accession number

GO:0005874

Gene class AmiGO h`p://www.geneontology.org/

[0-‐9][A-‐Z,0-‐9][A-‐Z,0-‐9][A-‐Z,0-‐9]

PDB ID 1TUB Protein, nucleic acid, or complex structure

PDB h`p://www.rcsb.org/

[2 or 3 le`ers or digits]

PDB ligand ID

CN2 Ligand PDB h`p://www.rcsb.org/



[up to 7 digits]-‐[2 digits]-‐[1 digit]

CAS registry number

64-‐86-‐8 Chemical structure

SciFinder h`ps://scifinder-‐cas-‐org.login.ezproxy.library.ualberta.ca/

[digits] PubChem CID (compound ID)

6167 Chemical structure

PubChem h`p://pubchem.ncbi.nlm.nih.gov/

ZINC[8 digits] OR [digits]

ZINC ID ZINC00621853

621853

Chemical structure

ZINC h`p://zinc.docking.org/

DB[5 digits] DrugBank accession number

DB01394 Drug (chemical structure)

DrugBank h`p://www.drugbank.ca/

Key File Formats for Sequences and Structures

•  Sequences – FASTA format .fasta .fst .txt!

•  Macromolecule structures – PDB format .pdb .ent!

Accessing Databases

•  Web interface

•  Query string e.g. http://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?! db=nucleotide&id=34222261&rettype=fasta&retmode=fasta!

•  Web services (SOAP)

•  FTP -‐> local copy

CAS SciFinder PubChem

DrugBank ZINC

Cheminforma6c Database Survey

h`ps://scifinder-‐cas-‐org.login.ezproxy.library.ualberta.ca/

h`p://pubchem.ncbi.nlm.nih.gov/

h`p://www.drugbank.ca/


DrugBank ZINC

h`p://zinc.docking.org/


>52 million organic compounds >61 million inorganic compounds

Physical property info

>27 million unique structures >23 million with 3d conforma=ons

Mostly organic, biologically interes=ng compounds

~4,800 drugs >1,350 FDA approved drugs

Includes drug target info


DrugBank ZINC

>13 million purchasable compounds

Ready to dock


N

O

N

O

SS

HO HN

O

N

O

NNH

SS

OH

5 Chaetocin structures from PubChem

CID 161591: no stereochemistry

N

O

N

O

SS

HO HN

O

N

O

NNH

SS

OH

CID 5390098: bad stereochemistry

N

O

N

O

SS

HO HN

O

N

O

NNH

SS

OH

CID 11657687: Natural product stereochemistry

N

O

N

O

SS

HO HN

O

N

O

NNH

SS

OH

CID 46191942: Enan=omer of natural product

N

O

N

O

SS

HO HN

O

N

O

NNH

SS

OH

CID 11563851: incomplete stereochemistry

Stereochemistry Issues

Other Cheminforma6c Issues

•  Tautomers / protona=on states? •  Salt forms?

•  Implicit or explicit hydrogens?

•  2D connec=vity only or 3D conforma=on?

•  Non-‐organic elements? – Many programs only handle: CHNOPS + halogens – But some drugs have B, Pt, Hg, As, …

CC(=O)N[C@H]1CCC2=CC!(=C(C(=C2C3=CC=C(C(=O)!C=C13)OC)OC)OC)OC

•  Isomeric SMILES –  Allows specifica=on of stereochemistry

•  Canonical SMILES –  Canonicaliza=on will generate a unique string for a molecule, regardless of

atom order –  Different programs will canonicalize differently

•  SMARTS –  Chemical pa`erns for searching or filtering

h`p://www.daylight.com/smiles/index.html

SMILES O H

N

O

O

OO

O

File formats

•  MDL Molfile .mol –  Allows a 3D conforma=on to be stored

•  SDF .sdf!– Wraps Molfile format; mul=ple structures; annota=ons

•  PDB .pdb .ent!–  Not the best for small molecules

Need to convert? -‐> Try OpenBabel h7p://openbabel.org/wiki/Main_Page

Pathway and Interac6on Databases

KEGG Pathways

NetPath

BioGRID


KEGG Pathways

NetPath

BioGRID

h`p://thebiogrid.org/

h`p://www.netpath.org/

h`p://www.genome.jp/kegg/


KEGG Pathways

NetPath

BioGRID

A repository for protein and gene interac=on data

345,620 interac=ons

Curated protein signal pathways in humans

20 pathways, 1,800 interac=ons

Manually drawn pathways of metabolism, signaling, and other biological processes

>300 pathways + organism specific versions

Pathway Formats

•  SBML .xml!– The Systems Biology Markup Language

h`p://sbml.org/Main_Page

•  Also check out the BioPAX format h`p://www.biopax.org/

Pathway Tools

•  libSBML h`p://sbml.org/SoUware/libSBML

•  Cell Designer h`p://www.celldesigner.org/

•  CytoScape h`p://www.cytoscape.org/

<?xml version="1.0" encoding="UTF-8"?><sbml level="2" version="3" xmlns="http://www.sbml.org/sbml/level2/version3">...<listOfSpecies> <species compartment="cytosol" id="ES" /> <species compartment="cytosol" id="P" /> <species compartment="cytosol" id="S" /> <species compartment="cytosol" id="E" /> </listOfSpecies> <listOfReactions> <reaction id="veq"> <listOfReactants> <speciesReference species="E"/> <speciesReference species="S"/> </listOfReactants> <listOfProducts> <speciesReference species="ES"/> </listOfProducts> <kineticLaw> <math xmlns="http://www.w3.org/1998/Math/MathML"> <apply> <times/> <ci>cytosol</ci>

KEGG: Pathways in Cancer

NetPath: EGFR1 pathway

Exercises

1.  What databases are these iden=fiers from? a.  3KYL b.  EZH2 c.  Q15910 d.  GO:0008017 e.  GI:8017 f.  A9145C

2.  Try finding the corresponding entries online

Exercise Answers

1.  What databases are these iden=fiers from? a.  3KYL -‐> PDB (a protein-‐RNA structure for telomerase

reverse transcriptase, cataly=c region) b.  EZH2 -‐> HGNC (a human gene for a histone lysine methyl

transferase) c.  Q15910 -‐> UniProt (a protein sequence for EZH2) d.  GO:0008017 -‐> AmiGO (microtubule binding gene

ontology) e.  GI:8017 -‐> GenBank (a DNA sequence from D.

melanogaster) f.  A9145C -‐> this one’s a trick: it’s a chemical compound;

you can look it up in PubChem with CID: 6438632

PharmaMatrixWorkshop2010 Bioinforma6cDatabasespwinter/Bioinformatic...manual#and#automa=c#...

Documents

Transcript of PharmaMatrixWorkshop2010 Bioinforma6cDatabasespwinter/Bioinformatic...manual#and#automa=c#...