Manolis KellismodENCODE analysis group

January 11, 2007

Part 1: Target identification: comparative vs. exprmt.

(really the topic for today)

Part 2: Target validation (optional)

Part 3: Motif discovery (optional)

Part 4: Enhancer identification (optional)

Part 1

Identifying targets using

comparative genomics

Evolutionary signatures of motif instances

• Allow for motif movements– Sequencing/alignment errors– Loss, movement, divergence

• Measure branch-length score– Sum evidence along branches– Close species little contribution

BLS: 25% Mef2:YTAWWWWTAR BLS: 83%

Motif confidence selects functional instancesTranscription factor motifs

Confidence

microRNA motifs

Confidence

Increasing BLS Increasing confidence

Confidence selects functional regions

Confidence selects in vivo bound sites

High sensitivity

Confidence selects positive strand

Increasing BLS Increasing confidence

Confidence selects functional regions

Initial regulatory network for an animal genome

• ChIP-grade quality– Similar functional

enrichment– High sens. High spec.

• Systems-level– 81% of Transc. Factors– 86% of microRNAs– 8k + 2k targets– 46k connections

• Lessons learned– Pre- and post- are

correlated (hihi/lolo)– Regulators are heavily

targeted, feedback loop

Network captures literature-supported connections

Network captures co-expression supported edges

Red = co-expressed

Grey = not co-expressed

Named = literature-supported

Bold = literature-supported46% of edges

are supported (P=10-3)

ChIP vs. conservation: similar power / complementary

• Together: best

complementary

• Bound but not conserved: reduced enrichmnt

Selects functional

• All-ChIP vs. All-cons: similar enr.

Similar power

• Cons-only vs. ChIP-all: similar

Additional sites

Part 2

Cool story of miRNA targets

for a new anti-sense miRNA

Surprise: miR-Anti-sense function

• A single miRNA locus transcribed from both strands• Both processed to mature miRNAs: mir-iab-4, miR-iab-4AS (anti-sense)• The two miRNAs show distinct expression domains (mutually exclusive)• The two show distinct Hox targets – another Hox master regulator

Surprise: miR-Anti-sense function

• Mis-expression of mir-iab-4S & AS: altereswings homeotic transform.

• Stronger phenotype for AS miRNA• Sense/anti-sense pairs as general

building blocks for miRNA regulation• 9 new anti-sense miRNAs in mouse

halterewing

wing

haltereSensory bristles

wing

w/bristles

sense Antisense

WT

No

te:

C,D

,E s

ame

mag

nif

icat

ion

Part 3 (optional)

Discovering motifs

Evolutionary signatures for regulatory motifs

• Individual motif instances are preferentially conserved• Measure conservation across entire genome

– Over thousands of motif instances Increased discovery power– Couple to rapid enumeration and rapid string search

De novo discovery of regulatory motifs

Known

engrailed

site

(footprint)

D.mel

D. ere

D. ana

D. pse.

5’-UTR 3’-UTR

D.mel CAGCT--AGCC-AACTCTCTAATTAGCGACTAAGTC-CAAGTCD.sim CAGCT--AGCC-AACTCTCTAATTAGCGACTAAGTC-CAAGTCD.sec CAGCT--AGCC-AACTCTCTAATTAGCGACTAAGTC-CAAGTCD.yak CAGC--TAGCC-AACTCTCTAATTAGCGACTAAGTC-CAAGTCD.ere CAGCGGTCGCCAAACTCTCTAATTAGCGACCAAGTC-CAAGTCD.ana CACTAGTTCCTAGGCACTCTAATTAGCAAGTTAGTCTCTAGAG ** * * *********** * **** * **

Consensus MCS Matches to known Expression enrichment Promoters Enhancers

1 CTAATTAAA 65.6 engrailed (en) 25.4 2

2 TTKCAATTAA 57.3 reversed-polarity (repo) 5.8 4.2

3 WATTRATTK 54.9 araucan (ara) 11.7 2.6

4 AAATTTATGCK 54.4 paired (prd) 4.5 16.5

5 GCAATAAA 51 ventral veins lacking (vvl) 13.2 0.3

6 DTAATTTRYNR 46.7 Ultrabithorax (Ubx) 16 3.3

7 TGATTAAT 45.7 apterous (ap) 7.1 1.7

8 YMATTAAAA 43.1 abdominal A (abd-A) 7 2.2

9 AAACNNGTT 41.2 20.1 4.3

10 RATTKAATT 40 3.9 0.7

11 GCACGTGT 39.5 fushi tarazu (ftz) 17.9

12 AACASCTG 38.8 broad-Z3 (br-Z3) 10.7

13 AATTRMATTA 38.2 19.5 1.2

14 TATGCWAAT 37.8 5.8 2

15 TAATTATG 37.5 Antennapedia (Antp) 14.1 5.4

16 CATNAATCA 36.9 1.8 1.7

17 TTACATAA 36.9 5.4

18 RTAAATCAA 36.3 3.2 2.8

19 AATKNMATTT 36 3.6 0

20 ATGTCAAHT 35.6 2.4 4.6

21 ATAAAYAAA 35.5 57.2 -0.5

22 YYAATCAAA 33.9 5.3 0.6

23 WTTTTATG 33.8 Abdominal B (Abd-B) 6.3 6

24 TTTYMATTA 33.6 extradenticle (exd) 6.7 1.7

25 TGTMAATA 33.2 8.9 1.6

26 TAAYGAG 33.1 4.7 2.7

27 AAAKTGA 32.9 7.6 0.3

28 AAANNAAA 32.9 449.7 0.8

29 RTAAWTTAT 32.9 gooseberry-neuro (gsb-n) 11 0.8

30 TTATTTAYR 32.9 Deformed (Dfd) 30.7

Power of evolutionary signatures for motif discovery

Ability to discover full dictionary of regulatory motifs de novo

Tissue-specific enrichment and clustering

• Infer candidate functions for novel motifs• Reveal ‘modules’ of co-operating motifs

Functional clusters emerge

Discovered motifs show positional biases

• May represent new core promoter elements• Show enrichment in distinct functional categories

Recognizing functional motifs in coding regions

• Challenge: – Overlapping selective pressures– Most ‘motifs’ from di-codon biases– Hundreds of motifs due to noise

• Solution: – Test each frame offset separately– Di-codon biases Frame biased– True motifs Frame unbiased

• Result: – Top 20 motifs 11 miRNA seeds– (before: 11 seeds in 200+ motifs)

miRNAs

Top motifs

Ability to distinguish overlapping pressures

Evidence of miRNA targeting in coding reg.

miRNA targeting in protein-coding regions

• MicroRNA seeds are specifically selected

• Coding & 3’UTRs show same conservation profile

Part 4 (optional)

Characterizing enhancers

Developmental enhancer identification in Drosophila

• Supported by tiling arrays and regulatory motifs (nucleotide resolution)

• Identify nearly all known enhancers (20 of 22 highly bound)

Bound in vivo. Conserved D/Tw/Sn motifs in 12 flies. Clear DV expression pattern (lacZ/end).

• Large number of novel enhancers (428 Dorsal/Twi/Sna). They validate!

Surprise 1: AP genes targeted by DV regulators

• Novel DorsoVentral enhancers in known AntPosterior genes– Bound in vivo by DV genes (by all three DV master regulators)– Show evolutionarily conserved motifs for all three DV factors– Yet, found in known AP genes, with clear AP expression patterns

Integration of DV and AP patterning networks

Surprise 2: Some silent genes show Pol II binding

• Distinct modes of Pol II occupancy– Active genes (27%): Pol II throughout the gene, transcribing– Repressed genes (37%): Pol II simply absent, no expression

• Third class (12%): Pol II found only at the TSS, stalled– Qualitatively different: abundantly bound, but strongly punctate– Genes not expressed: known repressed genes, confirmed by arrays– Enriched in development, neurogenesis, ectoderm, muscle differ.

• Hypothesis: Developmental genes poised for expression– Reminiscent of ‘bivalent’ K4/K27 domains in mammals

Active Repressed Poised

Surprise 3: Master regulators also bind downstream targets

• Abundant feed-forward loops in DV patterning

• Cooperation of master reg. & downstream reg.

Manolis Kellis - modENCODE analysis - summary

• Part 1: Target identification– Comp. vs. Expt: each has unique advantages– Bound & not conserved appear less functional!

• Part 2: Target validation (for anti-sense miRNA)– It’s nice when expected outcome comes true– Need more collaborations for target validation

• Part 3: Motif discovery– Methods for genome-wide motif discovery– Expect increased power in bound regions

• Part 4: Enhancer identification– Many new enhancers – with motifs & validation– AP / DV system cross-talk – expect dense network– PolII stalling: spatial dynamics matter

Who’s actually doing the work

Alex Stark

Collaborators: Targets Sushmita Roy @ UNM

iab-4AS Natascha Bushati, Steve Cohen @ EMBLJulius Brennecke, Greg Hannon @ CSHLCalvin Jan, David Bartel @ Whitehead

Enhancers Julia Zeitlinger, Rick Young @ WhiteheadRobert Zinzen, Mike Levine @ UC Berkeley

PouyaKheradpour

JuliaZeitlinger

Main contributors: