Effect of polymorphisms on transcriptional regulation in mice

1
Effect of polymorphisms on transcriptional
regulation in mice

• Preliminary study of SNPs in the promoter region
  of genes with strong cis-eQTL

Manish Anand Advisors Dr. Debraj GuhaThakurta,
Dr. Sun Kim ROSETTA INPHARMATICS, and INDIANA
UNIVERSITY 24th Sep, 2004
2
Overview

• Background
• Methods
• TFBS predictions genome-wide
• TF specific analysis
• Motifs in Promoter Regions
• Results
• Conclusions

3
Background
4
Central dogma
ZOOM IN
tRNA
transcription
DNA
rRNA
snRNA
translation
POLYPEPTIDE
mRNA
5
Transcription key steps

• Initiation
• Elongation
• Termination

DNA
6
Transcription key steps

• Initiation
• Elongation
• Termination

DNA
7
Transcription key steps

• Initiation
• Elongation
• Termination

DNA
DNA

RNA
8
Methods
9
Regulation of Genes

Transcription Factor
RNA polymerase
DNA
Coding region
Regulatory Element
10
(No Transcript)
11
SNP
Gene 1
SNP
Gene 2
SNP
Gene 3
12
SNPs
Upstream 1
Upstream 2
Upstream 3
Upstream 4
TF predictions
13
Selected Predictions
14
T
G
C
T
A
A
A
G
15
PATSER / MATCH SCORE 0.95
PATSER / MATCH SCORE 0.22
16
TF
0.95
TF
0.22
17
Gene Expression
Gene Expression
Mice images taken from Terry Speeds lecture How
many genes? Mapping mouse traits, January 20,
2004
18
Consequences of point mutations in the promoter
for the b-globin gene
(From T. Maniatis, S. Goodbourn, and J. A.
Fischer, Science 236, 1987, 1237.)
19
TFBS Predictions Analysis
20
General Summary
21
MATCHScore Difference Analysis
22
TF specific analysis
23
Motifs whose predicted binding sites are
disrupted (show score differences)
24
Motifs whose predicted binding sites are
disrupted (site drops)
25
MOTIF DISCOVERY
26
Novel MOTIFs in positive set
27
MOTIFs in negative set
28
Future Investigations
29
Future Investigations

• Mapping of real (experimentally verified) binding
  sites in TRANSFAC onto the genome
• Observing whether SNPs that disrupt these binding
  sites are in gene regions that show strong
  cis-acting eQTLs.
• Selection of several genes that show strong
  cis-eQTLs and have either real or predicted
  transcription factor binding sites disrupted in
  their upstream regions for experimental
  follow-up/validation with genotyping and
  measurement of expressions with different
  variants observed in the different strains of
  mice.
• Such experiments would help in identification of
  causal variants for expression changes in these
  genes.

30
Conclusions
31
Conclusions

• Amongst genes with cis-eQTLs, more genes have
  polymorphisms in their promoter regions
• Significantly more genes in the set with
  cis-acting eQTLs show score differences in
  putative TFBSs compared to control set
• Identification of specific TFs whose binding site
  predictions are disrupted in genes with
  cis-acting eQTLs
• Identification of novel DNA motifs around SNP
  regions in the positive set

32
Acknowledgements

• Indiana University
• Dr. Sun Kim
• Dr. Gary Wiggins
• Dr. Marty Siegel

• Rosetta Inpharmatics
• Dr. Debraj GuhaThakurta
• Dr. Eric Schadt
• Dr. John Lamb
• Dr. Stephen Edwards
• Dr. Barmak Modrek

Few images were taken from Basic Biology for
CS262 - OMKAR DESHPANDE, Stanford University