Discovering motifs in DNA sequences using AMADEUS

1
Discovering motifs in DNA sequences
usingAMADEUS
C. Linhart, Y. Halperin, R. Shamir
TAU workshop, May 07
2
Promoter Analysis Exteremely brief intro

• Transcription is regulated primarily by
  transcription factors (TFs) proteins that bind
  to DNA subsequences, called binding sites (BSs)
• TFBSs are located mainly (not always!) in the
  genes promoter the DNA sequence upstream the
  genes transcription start site (TSS)
• TFs can promote or repress transcription

TSS
3
Promoter Analysis (cont.)TFBS models

• The BSs of a particular TF share a common
  pattern, or motif, which is often modeled using
• Consensus string
• TASDAC (SC,G DA,G,T)
• Position weight matrix (PWM / PSSM)

gt Threshold 0.01 TACACC (0.06) TAGAGC
(0.06) TACAAT (0.015)
4
Promoter Analysis (cont.) Goals

• Reverse-engineer the transcriptional regulatory
  network find the TFs (and their BSs) that
  regulate the studied biological process
• Input A set of co-expressed genes
• Output Interesting motif(s)
• Known motifs PRIMA, ROVER,
• Novel motifs MEME, AlignACE,
• A group of co-occurring motifs
  cis-regulatory module (CRM) MITRA, CREME,

AMADEUS
5
Promoter Analysis (cont.) Challenges

• Why is it so difficult?
• BSs are short and degenerate (non-specific)
• Promoters are long complex (hard to model)
• Multiple BSs of several TFs
• Old (non-functional) BSs
• Other genetic/structural signals (e.g., GC
  content)
• Search space is huge
• 1510 (500 billion) consensus strings of length
  10
• 1Kbp promoter ? 2 strands ? 20K genes in human
  40 Mbps
• Which score to use - what makes a motif
  interesting?
• Enrichment over-representation w.r.t. BG model
• Location and/or strand bias
• Conservation across related species

6
(No Transcript)
7
Promoter Analysis (cont.) Status of motif
discovery tools

• Extant tools perform reasonably well for
• Finding known/novel motifs in organisms with
  short, simple promoters, e.g., yeast
• Identifying some of the known motifs in complex
  species, e.g., TFs whose BSs are usually close to
  the TSS
• but often fail in other cases!
• Each tool is custom-built for a specific target
  score
• Comparison of tools Tompa et al. 05

8
AMADEUS
A Motif Algorithm for Detecting Enrichment in
mUltiple Species

• Research platform
• Extensible add new algs, scores, motif models
• Flexible control params, algs, scores of
  execution
• Experimental tool
• Sensitive find subtle signals
• Efficient handle huge amount of data
• Informative show lots of info on motifs
• User-friendly nice GUI

9
Main features I/O

• Input
• Target set in one or more species (or expression
  data)
• Sequence region (promoter, 1st intron, 3 UTR, )
  
• Various parameters
• Output
• Non-redundant set of motifs
• Rich info per output motif
• Graphical motif logo
• Multiple scores combined p-value
• Similarity to known TFBS models
• List of target genes
• BS localization graph

10
Main features alg.

• Algorithm Multiple refinement phases
• Each phase receives best candidates of previous
  phase,and refines them (e.g., uses a more
  complex motif model)
• First phases are simple and fast (e.g., try all
  k-mers) Last phases are more complex (e.g.,
  optimize PWM using EM)

11
Main features scores

• Motif scores
• User selects scores to use, a subset of
• Target-set over/under-representation
• Hyper-geometric
• GC-contentlength binned binomial
• Localization
• Strand bias
• (additional scores for expression data)
• Scores are combined into a single p-value
• Dont assume specific models for distribution of
  BSs
• (most tools use a statistical model, e.g.,
  Markov Model, to describe the promoter sequences)

12
Main features misc.

• GUI
• Control all parameters
• Save/load parameters from file
• Save textual and graphical output to file
• TFBS viewer
• Other
• Ignore redundant sequences (with identical
  subsequence)
• Applicable to multiple genome-scale promoter
  sequences
• Bootstrapping Empirical p-value estimation using
  random target sets / shuffled data
• Execution modes GUI , batch
• Interoperability Java application

13
Combining p-values

• Each motif receives p-values from various sources
  (several scores, multiple species) p1,p2,,pn
• We combine them into a single p-value p
• p Prob f1? f2?? fn ? p1? p2? ? pn fi
  U0,1
• Denote ? p1?p2??pn
• p 1 - ? ? ?(ln 1/?)i/i! , i0,,n-1
• Also developed a weighted version when each
  p-value has a different weight

14
Results I E2F targets Ren et al. 02
E2F
NF-Y
15
Results I G2 G2/M phases of human cell cycle
Whitfield et al. 02
CHR (not in TRANSFAC)
NF-Y
Module CHR and NF-Y motifs co-occur
(Module was reported in Linhart et al., 05,
Tabach et al. 05)
16
Results II Localized humanmouse motifs

• Input
• All human mouse promoters (2 x 21K)
• Region -500100 (w.r.t. TSS)
• Total sequence length 26 Mbps
• No target-set / expression data
• Score localization
• Results
• Recovered known TFs Sp1, NF-Y, Elk-1, TATA,
  Nrf-1, ATF/CREB, cMyc, RFX1
• Recovered the splice donor site
• Identified several novel motifs

17
Results III Yeast TF target sets Harbison et
al. 04
Number of motifs successfully recovered by
AMADEUS and 3 other motif finders (similarity was
computed w.r.t. the motifs reported by Harbison
et al. the 3 top scoring motifs found by each
alg were considered)
18
Questions?