TargetScan

http://www.targetscan.org/

TargetScanPrediction of microRNA targets

Release 6.2: June 2012

By Krittika Dummunee 5510220011 Bundit Boonyarit 5410210278

http://upload.wikimedia.org/wikipedia/commons/thumb/a/a7/MiRNA.svg/2000px-MiRNA.svg.png

microRNAs (miRNAs) are small non-coding RNAs about 22 nucleotides in length that play an important role in posttranscriptional regulation of target genes both in plant and animal cells.

2


miRNAs are produced by the enzyme Dicer from endogenous stem-loop RNA molecules, and are implicated in the control of several biological processes such as differentiation, cell proliferation and developmental timing.

3

=


- Nucleotide 1 has an “A” across from it - Seed region (nucleotides 2-8) perfect base pairing - Nucleotide 9 has an “A” or “U” across from it - Nucleotide 13-16 good base paring

4

miRNA-mRNA interaction

The importance of the miRNA regulatory pathways is underscored by the impressive list of diseases which have recently been found to be associated with abnormal miRNA expression.

Future Directions for miRNA

miRNA may also be involved in other processes besides translational gene silencing. Currently there are hints of this, because mature mammalian miRNAs can be imported into the nucleus [17] and secreted from the cell [18]. These results suggest that miRNA may regulate transcription or paracrine signaling. Unlike siRNA, miRNA is endogenous, and therefore has the potential to enhance the understanding of the regulation of particular genes. In addition, miRNA is now touted as an additional layer of gene regulation, which can be dysregulated in diseases. Currently the study of miRNAs requires large scale arrays, since few miRNA targets are experimentally confirmed and individual miRNAs may have overlapping functions. The relative lack of attention devoted to miRNA will change in the future, as scientists realize that their favorite gene may have an additional layer of regulation never touched upon. While siRNA is merely an important tool, miRNA is evolving into a whole new field of research.

Table 1: Comparison of miRNA and siRNA.

miRNA

siRNA

Where Found?

Endogenous

Exogenous

Length

19-25 nt

19-21 nt

Target Recog

ImperfectMatchExactMatch

Mechanism

TranslationalRepression

mRNACleavage

miRNA

miR-15/miR-16miR-1

miR-146 miR-520hmiR-106amiR-let7miR-155miR-223miR-208

Function

ApoptosisCardiac Arrhythmia

Bacterial Infectious Response;TLR-NFκB

Stem Cell DifferentiationCancer PathogenesisCell Cycle RegulationAdaptive Immunity

Granulocyte RegulationStress Response (Heart)

Target

Bcl2GJA1/KCNJ2

IRAK1/TRAF6 ABCG2

Rb1Multiple

---

Reference

[19][20]

[21] [22][23][25][26][27][24]

Table 2: Examples of miRNA Functions & Relevance of miRNA to Human Biology.Age-Related Diseases Evidence is accumulating that many age-related diseases are associated with a decreased control of cell signaling that occurs in mid-life [25]. The miRNA control of such systems as the cell cycle, DNA repair, oxidative stress responses and apoptosis, has been shown to become abnormally expressed in mid-life. It is highly likely that continued research will reveal important associations with the aging process, and may lead to therapeu-tics that can improve the quality of life. Heart Disease Two heart-specific miRNAs were deleted in mouse models resulting in abnormal heart development in a large proportion of the offspring [25]. While these lethal effects were expected, other studies show a more subtle role for miRNA in the heart. When miR-208 was eliminated, the mice appeared normal. Defects were revealed only when their hearts were stressed. These results show that comprehensive miRNA studies may be valuable in the diagnosis of heart disease. Neurological Diseases Numerous reports have demonstrated the role of miRNAs in neural development. Evidence for a role in Parkinsons disease comes from animal model studies published last year, showing that loss of miRNAs may be involved in the development and progession of the disease. In cell culture experiments, transfer of small RNA fragments partially preserved miRNA deficient nerve cells [25]. While these results and others point to an important role for miRNA in neurodegenerative disorders, much more work is needed to delineate the exact role of miRNAs in this important area. Immune Function Disorders Recent miRNA deletion studies have revealed a central role in the regulation of the immune response. The deletion of miRNA-155 impaired T and B cell differentiation in germinal centers, and greatly decreased antibody and cytokine production [24]. Two additional studies deleting miRNA-181 and 223 were found to control T cell response and granulocyte production, respectively [25]. As more roles for miRNAs in the immune response are found, the list of immune function disorders with a miRNA component is certain to expand also.

Complete System for miRNA Research from SABiosciences

SABiosciences’ RT2 miRNA PCR Arrays & qPCR Assays generate high-quality and genome-wide miRNA expression data with nothing more than a simple RT-PCR protocol. Our patented miRNA technology ingeniously integrates a universal tailing & reverse transcription reaction specific for miRNA with the accurate expression level measurement of distinct miRNA sequences that may only differ by a single nucleotide base. With this technology, you can easily get a comprehensive survey of miRNA expression in your cell line or tissue of interest.

SABiosciences’ complete miRNA PCR System includes: RT2 miRNA Arrays and Assays

RT2 miRNA First Strand Kit RT2 SYBR Green PCR Master Mix

What the System Offers:

Detecting every miRNA across the entire genome in a specific and sensitive way is a very challenging technology task. Many miRNA family members and otherwise distinct miRNA species have very similar sequences. Moreover, other RNA species such as snRNA, tRNA, mRNA, and rRNA can cause non-specific amplification, making the specific analysis of mature miRNA even more problematic. With SABiosciences’ complete miRNA PCR System & expression analysis system, these problems are solved.

RT2 miRNA PCR Arrays and Assays dramatically improve the specificity through patent pending primer design & proprietary reverse transcription chemistry. Our miRNA PCR Arrays include built-in control elements to insure the quality of your experimental data. The free data analysis software takes your raw threshold cycle data and automatically generates figures and tables ready for publication. With the RT2 miRNA PCR Assay and Arrays, you can expect:

Sensitivity: As little as 0.5 µg total RNA needed

Multi-Sequence Flexibility: Analyze one to 376 sequences simultaneously

Simplicity: As easy as a real-time PCR experiment

15

FEATURED PATHWAY REVIEW: MicroRNA: Why Study It & How

©2008 www.SABiosciences.com SABiosciencesTM

http://www.sabiosciences.com/pathwaymagazine/pathways7/microrna.php 5

Figure 4. MicroRNA targetinga, Expanded repertoire of seed-matched sites preferentially conserved in nematode 3′UTRs.Sites conserved only marginally above chance are above the dashed line. Watson-Crick-matched residues, blue or black; residues independent of the miRNA sequence, red. b,Density of miRNA sites conserved above background, combining all site types at themaximally sensitive cutoff. Error bars, one standard deviation (calculated by repeating theanalysis for each site type 50 times, each time using a different cohort of control sequencesthat matched the properties of the miRNA sequences18). c, Relative strength of miRNA sitetypes across clades. Within each clade, two species of comparable divergence were selected.For each miRNA site type, the fraction of sites conserved above background in the twospecies was normalized to that of the 8mer-A1 (shown in parentheses). d, Enrichment of8mer-A1 3′UTR sites above expectation based on dinucleotide content. Error bars, onestandard deviation, derived as in (b). e, Relationship between 3′UTR length and siteenrichment. Site enrichment is ploted for 3′UTRs of the indicated species sorted by lengthinto ten equally sized bins.

Jan et al. Page 15

Nature. Author manuscript; available in PMC 2011 July 6.

NIH

-PA

Author M

anuscriptN

IH-P

A A

uthor Manuscript

NIH

-PA

Author M

anuscript

miRNA-mRNA interaction

Type of miRNA-mRNA interaction

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3057491/pdf/nihms249209.pdf 6

TargetScan

http://www.targetscan.org/

TargetScanPrediction of microRNA targets

Release 6.2: June 2012

7


Jan et al. Page 15


NIH

-PA

Author M

anuscriptN

IH-P

A A

uthor Manuscript

NIH

-PA

Author M

anuscript

TargetScan

Type of miRNA-mRNA interaction

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3057491/pdf/nihms249209.pdf 8

TargetScan

Interface

9

TargetScan

UsageSelect a species

10

TargetScan

Usage Enter a human Entrez Gene symbol

“Calcitonin receptor”

11

TargetScanTarget gene Gene name Conserved sites

total 8mer 7mer-m8 7mer-1ASMG1 PI-3-kinase-related kinase SMG-1 2 1 0 1SSR3 signal sequence receptor, gamma (translocon-associated protein gamma) 2 1 0 1TNNI1 troponin I type 1 (skeletal, slow) 2 1 0 1APLN apelin 1 1 0 0BCL11B B-cell CLL/lymphoma 11B (zinc finger protein) 1 1 0 0C1orf43 chromosome 1 open reading frame 43 1 1 0 0CNTNAP1 contactin associated protein 1 1 1 0 0CRTC2 CREB regulated transcription coactivator 2 1 1 0 0CSF1R colony stimulating factor 1 receptor 1 1 0 0DDX54 DEAD (Asp-Glu-Ala-Asp) box polypeptide 54 1 1 0 0FLJ11151 hypothetical protein FLJ11151 1 1 0 0GBA2 glucosidase, beta (bile acid) 2 1 1 0 0GOSR1 golgi SNAP receptor complex member 1 1 1 0 0IGF2 insulin-like growth factor 2 (somatomedin A) 1 1 0 0IKZF2 IKAROS family zinc finger 2 (Helios) 1 1 0 0LEP leptin 1 1 0 0LOXL3 lysyl oxidase-like 3 1 1 0 0MEOX1 mesenchyme homeobox 1 1 1 0 0MEX3A mex-3 homolog A (C. elegans) 1 1 0 0MPP2 membrane protein, palmitoylated 2 (MAGUK p55 subfamily member 2) 1 1 0 0MRFAP1L1 Morf4 family associated protein 1-like 1 1 1 0 0MRPL11 mitochondrial ribosomal protein L11 1 1 0 0MYADM myeloid-associated differentiation marker 1 1 0 0OSM oncostatin M 1 1 0 0PDLIM2 PDZ and LIM domain 2 (mystique) 1 1 0 0PIB5PA phosphatidylinositol (4,5) bisphosphate 5-phosphatase, A 1 1 0 0PPP2R5D protein phosphatase 2, regulatory subunit B', delta isoform 1 1 0 0PURB purine-rich element binding protein B 1 1 0 0RAB11FIP1 RAB11 family interacting protein 1 (class I) 1 1 0 0RAG1 recombination activating gene 1 1 1 0 0RP1L1 retinitis pigmentosa 1-like 1 1 1 0 0

SCRT2 scratch homolog 2, zinc finger protein (Drosophila) 1 1 0 0

SEMA3E sema domain, immunoglobulin domain (Ig), (semaphorin) 3E 1 1 0 0

SETX senataxin 1 1 0 0

SLC30A3 solute carrier family 30 (zinc transporter), member 3 1 1 0 0



SMURF1 SMAD specific E3 ubiquitin protein ligase 1 1 1 0 0

SUSD1 sushi domain containing 1 1 1 0 0

TLR4 toll-like receptor 4 1 1 0 0

TNPO1 transportin 1 1 1 0 0

USP1 ubiquitin specific peptidase 1 1 1 0 0

ZNF24 zinc finger protein 24 1 1 0 0



ATRN attractin 1 0 1 0

BTG1 B-cell translocation gene 1, anti-proliferative 1 0 1 0

CALN1 calneuron 1 1 0 1 0

CD28 CD28 molecule 1 0 1 0

DYRK1A dual-specificity tyrosine-(Y)-phosphorylation regulated kinase 1A 1 0 1 0

FZD4 frizzled homolog 4 (Drosophila) 1 0 1 0

GABRE gamma-aminobutyric acid (GABA) A receptor, epsilon 1 0 1 0

IMPDH1 IMP (inosine monophosphate) dehydrogenase 1 1 0 1 0

KCNE3 potassium voltage-gated channel, Isk-related family, member 3 1 0 1 0

KRT76 keratin 76 1 0 1 0

MARK4 MAP/microtubule affinity-regulating kinase 4 1 0 1 0

NAV1 neuron navigator 1 1 0 1 0

PBX1 pre-B-cell leukemia homeobox 1 1 0 1 0

SDC1 syndecan 1 1 0 1 0

SLC45A3 solute carrier family 45, member 3 1 0 1 0

SUV39H1 suppressor of variegation 3-9 homolog 1 (Drosophila) 1 0 1 0

TERF2 telomeric repeat binding factor 2 1 0 1 0

TMEM132B transmembrane protein 132B 1 0 1 0

ESR1 estrogen receptor 1 1 0 0 1

http://www.plosone.org/article/fetchSingleRepresentation.action?uri=info:doi/10.1371/journal.pone.0097200.s001. 12

TargetScan

ARFIP2 ADP-ribosylation factor interacting protein 2 (arfaptin 2) 1 0 0 1C10orf104 chromosome 10 open reading frame 104 1 0 0 1C15orf55 chromosome 15 open reading frame 55 1 0 0 1C3orf10 chromosome 3 open reading frame 10 1 0 0 1CALCR calcitonin receptor 1 0 0 1CNNM3 cyclin M3 1 0 0 1DDX3X DEAD (Asp-Glu-Ala-Asp) box polypeptide 3, X-linked 1 0 0 1DENND1A DENN/MADD domain containing 1A 1 0 0 1EIF5A2 eukaryotic translation initiation factor 5A2 1 0 0 1ELMO2 engulfment and cell motility 2 1 0 0 1FOXA1 forkhead box A1 1 0 0 1GATAD2B GATA zinc finger domain containing 2B 1 0 0 1ITCH itchy E3 ubiquitin protein ligase homolog (mouse) 1 0 0 1LIMD1 LIM domains containing 1 1 0 0 1LYRM1 LYR motif containing 1 1 0 0 1MRFAP1 Mof4 family associated protein 1 1 0 0 1NPTXR neuronal pentraxin receptor 1 0 0 1ORMDL3 ORM1-like 3 (S. cerevisiae) 1 0 0 1PIP4K2B phosphatidylinositol-5-phosphate 4-kinase, type II, beta 1 0 0 1PLAGL2 pleiomorphic adenoma gene-like 2 1 0 0 1PLEKHH3 pleckstrin homology domain containing, family H member 3 1 0 0 1PML promyelocytic leukemia 1 0 0 1RAB6A RAB6A, member RAS oncogene family 1 0 0 1RC3H1 ring finger and CCCH-type zinc finger domains 1 1 0 0 1RECQL5 RecQ protein-like 5 1 0 0 1SH3PXD2A SH3 and PX domains 2A 1 0 0 1SH3TC2 SH3 domain and tetratricopeptide repeats 2 1 0 0 1SHANK2 SH3 and multiple ankyrin repeat domains 2 1 0 0 1SLC10A7 solute carrier family 10, member 7 1 0 0 1SLC5A12 solute carrier family 5 (sodium/glucose cotransporter), member 12 1 0 0 1SMYD3 SET and MYND domain containing 3 1 0 0 1SNX27 sorting nexin family member 27 1 0 0 1SRGAP3 SLIT-ROBO Rho GTPase activating protein 3 1 0 0 1

STK4 serine/threonine kinase 4 1 0 0 1SYNGR1 synaptogyrin 1 1 0 0 1SYS1 SYS1 Golgi-localized integral membrane protein homolog (S. cerevisiae) 1 0 0 1TNFAIP1 tumor necrosis factor, alpha-induced protein 1 (endothelial) 1 0 0 1XPO4 exportin 4 1 0 0 1ZDHHC16 zinc finger, DHHC-type containing 16 1 0 0 1ZNF648 zinc finger protein 648 1 0 0 1

http://www.plosone.org/article/fetchSingleRepresentation.action?uri=info:doi/10.1371/journal.pone.0097200.s001. 13

TargetScan

UsageSubmit

14

TargetScan

Results

15

TargetScan

16

Results

TargetScan

17

Results

TargetScan

18

Results

TargetScan

19

Results

TargetScan


Jan et al. Page 15


NIH

-PA

Author M

anuscriptN

IH-P

A A

uthor Manuscript

NIH

-PA

Author M

anuscript

20

Results

TargetScan

“The site-type contribution reflects the average contribution of each site type. A more negative score is associated with a more favorable site (Grimson et al., 2007).”

21

Results

TargetScan

“The 3' pairing contribution reflects consequential miRNA-target complementarity outside the seed region. A more negative score is associated with a more favorable site (Grimson et al., 2007).”

22

Results

TargetScan

“The local AU content reflects the transcript AU content 30 nt upstream and downstream of predicted site. A more negative score is associated with a more favorable site (Grimson et al., 2007).”

23

Results

TargetScan

“The position contribution reflects the distance to the nearest end of the annotated UTR of the target gene. A more negative score is associated with a more favorable site (Grimson et al., 2007).”

24

Results

TargetScan

“The target site abundance contribution (TA) to context+ score reflects the abundance of target sites of a miRNA family in the set of distinct 3' UTRs (Garcia et al., 2011). A more negative score

is associated with a lower abundance of the miRNA target site in the set of 3' UTRs.”

25

Results

TargetScan

“The seed-pairing stability contribution (SPS) to context+ score reflects the stability of of a miRNA-target duplex, which is a function of the concentration of (A+U) in the seed region

(Garcia et al., 2011). A more negative score is associated with weaker seed-pairing stability.”

26

Results

TargetScan

“The context+ score for a specific site is the sum of the contribution of these six features, calculated as in Garcia et al., 2011: site-type contribution 3' pairing contribution local AU contribution position contribution TA (target site abundance) contribution* SPS (seed-pairing stability) contribution*” 27

Results

TargetScan

“The context+ score percentile rank is the percentage of sites for this miRNA with a less favorable context+ score.”

28

Results

TargetScan

“The conserved branch length score (Friedman et al., 2009) is the sum of phylogenetic branch lengths between species that contain a site. To help control for individual UTR conservation, 3' UTRs were separated by conservation rate into ten equally sized bins, and a unique set of branch lengths based on 3' UTR sequence alignments was constructed for each bin. Site conservation is defined by conserved branch length, with each site type having a different threshold for conservation: 8mer: 0.8 7mer-m8: 1.3 7mer-1A: 1.6” 29

Results

TargetScan

“PCT, the probability of conserved targeting as described in Friedman et al., 2009, has been calculated for all highly conserved miRNA families.”

30

Results

TargetScan

31

Results

Advantages & Disadvantages96 C

urrent Molecular M

edicine, 2011, Vol. 11, N

o. 2 W

itkos et al.

Features Experimental evaluation results Assessment

Sethupathy et al. 2006

Baek et al. 2008

Alexiou et al. 2009

Target prediction algorithm

Parameters contributing to the final score

Cross-species conservation

sensitivity1 log2-fold change2 precision3 sensitivity4

Advantages Disadvantages

Ref

eren

ce

miRanda complementarity and free energy

binding

conservation filter is used 49% 0.14 29% 20%

- beneficial for prediction sites with imperfect binding

within seed region

- low precision, too many false positives [56]

TargetScan

seed match, 3’ complementarity local AU content

and position contribution5

given scoring for each result 21% 0.326 51% 12%

- many parameters in-cluded in target scoring

- final score correlates with protein downregulation

- sites with poor seed pairing are omitted [38]

Target-ScanS seed match type

only conservative sites are consid-

ered 48% – 49% 8%

- simple tool for search of conserved sites with strin-

gent seed pairing

- underestimate miR-NAs with multiple

target sites [38]

PicTar binding energy, complementarity and conservation

required pairing at conserved posi-

tions 48% 0.26 49% 10% - miRNAs with multiple

alignments are favored

- does not predict non-conservative

sites

[42, 57]

DIANA-microT

free energy bind-ing and comple-

mentarity

dataset of con-served UTRs

among human and mouse is used

10% – 48% 12%

- SNR ratio and probability given for each target site

- possibility of using own miRNA sequence as an

input

- some miRNAs with multiple target sites

may be omitted [36]

PITA target site accessibility

energy

user-defined

cut-off level – 0.046 26% 6%

- the secondary structure of 3’UTR is considered for

miRNA interaction

- low efficiency com-pared to other algo-

rithms [46]

Rna22 pattern recognition and folding energy not included – 0.09 24% 6%

- allows to identify

sites targeted by yet-undiscovered miRNAs

- low efficiency com-pared to other algo-

rithms [58]

1percentage of experimentally supported miRNA-target gene interactions predicted (used TarBase records for which a direct miRNA effect was examined). 2average protein depression of genes predicted by the algorithm to be miR-223 targets. 3proportion of correctly predicted target miRNAs to total predicted miRNA-mRNA interactions (data obtained from proteomic analyses carried out by Selbach et al.). 4proportion of correctly predicted target miRNAs to total correct miRNA-mRNA interactions (data obtained from proteomic analyses carried out by Selbach et al.). 5position contribution parameter promotes sites close to the 3’UTR ends. 6the final scoring correlates with the level of protein downregulation.

microRNA Target Prediction Current Molecular Medicine, 2011, Vol. 11, No. 2 95

Fig. (1). Types of miRNA-mRNA interactions. Different classes of miRNA target sites are presented in a schematic way. Vertical dashes represent single Watson-Crick pairing. Nucleotides involved in binding have been arbitrarily defined to depict positions of required complementarity between miRNA and mRNA. Seed regions of miRNAs are marked by red color and the adenine at binding position 1 by green. Interactions between mRNA and the 3’ end of miRNA have not been shown because they are sequence-dependent and do not significantly contribute to the miRNA downregulation effect. In the case of 3’-suppelmentary and 3’-compensatory sites two regions of pairing (base pairs colored in blue) force middle mismatches to form a loop structure. Additionally, features of particular site types have been listed.

miRNA TARGET PREDICTION ALGORITHMS Many different algorithms have been developed for

prediction of miRNA-mRNA interactions. The rules for targeting transcripts by miRNAs have not been fully examined yet and are based mainly on experimentally validated miRNA-mRNA interactions [49, 50] that are only a slice of possibly existing in vivo. This situation led to the development of a variety of approaches to miRNA target prediction. The available algorithms have been extensively discussed by others [51-54]. Moreover, they were recently reviewed by Yue et al. [55] with the focus on their bioinformatical, mathematical and statistical aspects. The available algorithms can be classified into two categories established on the basis of the use or non-use of conservation comparison, a feature that influence greatly an outcome list of targets by narrowing the results [1, 33]. The algorithms based on conservation criteria are for example the following: miRanda [56], PicTar [42, 57], TargetScan [38], DIANA-microT [36];

while PITA [46] and rna22 [58] belong to the algorithms using other parameters, such as free energy of binding or secondary structures of 3’UTRs that can promote or prevent miRNA binding. Since all these algorithms were successfully used to predict miRNA targets in mammals we describe them in more detail below. Additionally, to facilitate the assessment of these algorithms, we summarize their performance and characteristic features (Table 1).

miRanda The miRanda algorithm [56] is based on a

comparison of miRNAs complementarity to 3’UTR regions. The binding energy of the duplex structure, evolutionary conservation of the whole target site and its position within 3’UTR are calculated and account for a final result which is a weighted sum of match and mismatch scores for base pairs and gap penalties. There is one wobble pairing allowed in the seed region

32

TargetScan

Science

Transcript of TargetScan