Differential location analysis
Download
Report
Transcript Differential location analysis
Characterization of transcriptional
responses to environmental stress
by differential location analysis
Junguk Hur
School of Informatics &
Center for Genomics and Bioinformatics
Indiana University
Bloomington, IN
OVERVIEW
Overview
Background / Motivation
Location Analysis / Differential Binding
TF Response Classification
Comparison with Microarray
Conclusion / Future Work
05-26-2006
Capstone Presentation
Junguk Hur
2
Central Dogma
Overview
Background
http://campus.queens.edu/faculty/jannr/bio103/tests/TEST2Help.htm
05-26-2006
Capstone Presentation
Junguk Hur
3
Transcriptional Regulation
Overview
Background
Albert et al., Molecular Cell Biology of the Cell
05-26-2006
Capstone Presentation
Junguk Hur
4
Transcriptional Regulation
Overview
Background
Transcription Factors
05-26-2006
Capstone Presentation
Junguk Hur
5
Response to Environment
Overview
Background
Broach et al. Curr Opin Microbiol 2004
05-26-2006
Capstone Presentation
Junguk Hur
6
Previous Studies - Expression
Overview
Background
High-throughput DNA Microarray – mRNA expression
http://www.fao.org/DOCREP/003/X6884E/x6884e03.htm
05-26-2006
Capstone Presentation
Junguk Hur
7
Previous Studies - Expression
Overview
Background
Literature : Gene Expression Microarray Stress
•
•
•
•
•
•
•
Exploring the metabolic and genetic control of gene expression on a
genomic scale, DeRisi JL, et al. (1997) Science
Genomic expression programs in the response of yeast cells to
environmental changes, Gasch AP, et al. (2000) Mol Biol Cell
Global and specific translational regulation in the genomic response of
Saccharomyces cerevisiae to a rapid transfer from a fermentable to a
nonfermentable carbon source, Kuhn KM, et al. (2001) Mol Cell Biol
Role of thioredoxin reductase in the Yap1p-dependent response to
oxidative stress in Saccharomyces cerevisiae, Carmel-Harel O, et al.
(2001) Mol Microbiol
Transcriptional Remodeling in Response to Iron Deprivation in
Saccharomyces cerevisiae, Shakoury-Elizeh M, et al. (2004). Mol Biol
Cell
Transcriptional response of steady-state yeast cultures to transient
perturbations in carbon source, Ronen M and Botstein D (2005) Proc
Natl Acad Sci
About 900 publications
05-26-2006
Capstone Presentation
Junguk Hur
8
Previous Studies – ChIP-Chip
Overview
Background
Which genes are directly regulated by TFs?
(Environment specific use of regulatory code)
Harbison et al. Nature 2004
05-26-2006
Capstone Presentation
Junguk Hur
9
Motivation and Goal
Overview
Background
Limitation of Microarray data for understanding
regulatory system upon environmental change
Previous qualitative analysis of ChIP-Chip
Integration of heterogeneous data
ChIP-Chip (direct regulation) +
Microarray (direct/indirect regulation)
Quantitative analysis of TF binding
Better understanding of differential responses of
transcriptional regulatory system via differential
location analysis
05-26-2006
Capstone Presentation
Junguk Hur
10
Location Analysis
Overview
Background
Location
Analysis
Genome-wide Location Analysis : ChIP-on-Chip Exp.
• In vivo assay based on
• ChIP (Chromatin Immuno-Precipitation)
• High-throughput array experiment
Where and how strongly TF binds to
05-26-2006
Capstone Presentation
Junguk Hur
11
Differential Location Analysis
Overview
Background
Location
Analysis
• Harbison et al. Nature 2004
• Saccharomyces cerevisiae (budding yeast)
• 204 TFs in 14 conditions (352 experiments)
• Genome-wide location data (11,000 interactions)
05-26-2006
Capstone Presentation
Junguk Hur
12
Data Sets - I
Overview
Background
Location
Analysis
ChIP-on-Chip
204 Yeast Transcription Factors (TFs)1
1 rich medium condition + 13 stress conditions
6540 genes & corresponding promoter sequences.
Microarray
Stress Condition
Concentration
Time point
Array
Ref.
H2O2 – oxidation
0.3 mM
20 mins
cDNA
2
H2O2 – oxidation
0.32 mM
20 mins
cDNA
3
H2O2 - oxidation
0.4 mM
20 mins
oligo
4
SM (Sulfometuron methyl)
– AA starvation
0.2 g/ml
15 mins
cDNA
5
SM (Sulfometuron methyl)
– AA starvation
0.2 g/ml
240 mins
cDNA
5
05-26-2006
Capstone Presentation
Junguk Hur
13
Data Sets - II
Overview
Background
Location
Analysis
ChIP-on-Chip Data Preprocessing
P-value and ratio data from Harbison et al.
‘NaN’ point removal
Ratio value below 1 set to 1
Stress conditions
13 stress conditions 147 TF-cond pairs
05-26-2006
Capstone Presentation
Junguk Hur
14
Differential Binding Ratio
Overview
Background
Location
Analysis
Rich medium
Stress Condition
Differential
Binding
Increased binding in stress condition
Decreased binding in stress condition
How different the binding ratio btw different conditions???
05-26-2006
Capstone Presentation
Junguk Hur
15
Differential Binding Measure
Overview
Background
Location
Analysis
Differential
Binding
Differential binding is represented as Pik by using
ChIP-chip binding ratio data between stress condition
(A) and rich medium (B).
Rich medium
Ai, Bi: binding ratio of TF k to
regulated region i
under stress and rich medium culture
Stress Condition condition respectively
-1 KPik 1
05-26-2006
Capstone Presentation
Pi > 0
UP
P iK < 0
DOWN
Junguk Hur
16
Distribution of Pik - I
Overview
Background
Location
Analysis
Differential
Binding
Ex) FHL1
YPD (Rich Medium) vs SM (Amino acid starvation)
All ChIP-Chip Binding Data
Normal distribution
05-26-2006
Capstone Presentation
High-confidence Binding Data
(p0.001)
Skewed to negative direction
Junguk Hur
17
Distribution of Pik - II
Overview
Background
-1 Pik 1
0.5 |Pik| 1
Up vs Down
HighUp vs HighDown
210
130
180
110
Num of Down regions
Differential
Binding
• Number of UP/DOWN differentially bound regions
(genes) for each TF-cond pair
• High-confidence data point only (p<0.001)
Num of Down regions
Location
Analysis
150
120
90
60
30
30
60
90
120
150
180
210
Num of Up regions
05-26-2006
70
50
30
10
0
-30
0
-30
90
Capstone Presentation
-10
10
30
50
70
90
110
130
Num of Up regions
Junguk Hur
18
Chi-Square Test - I
Overview
Background
Location
Analysis
Differential
Binding
* Chi-Square tests for Pik of 147 TF-condition pairs to
check their distributions (Normal or skewed)
* R-package & Perl scripts
* P<0.05 Skewed distribution
All ChIP-Chip Binding Data
High-confidence Binding Data
(p0.001)
Normal Dist. (p=0.97)
Skewed Dist. (p=0.00003)
05-26-2006
Capstone Presentation
Junguk Hur
19
Chi-Square Test - II
Overview
Background
Location
Analysis
Skewed Distribution
Chi-square test : p < 0.05
105 TF-Cond pairs (out of 147) : 70%
Differential
Binding
05-26-2006
Capstone Presentation
Junguk Hur
20
Classification of TFs - I
Overview
Background
Location
Analysis
Differential
Binding
TF response
classification
Depending on patterns of the skewed
pattern , 105 TF-condition pairs have been
grouped into three categories.
UP
: More than 2/3 are in ‘+’ direction
DOWN
: More than 2/3 are in ‘–’ direction
BOTH
: Similar proportions of ‘+’ and ’-’
05-26-2006
Capstone Presentation
Junguk Hur
21
Classification of TFs - II
Overview
Background
Location
Analysis
UP : YAP5_H2O2Hi
BOTH : HPO4_Pi
DOWN : SFP1_H2O2Lo
Differential
Binding
TF response
classification
05-26-2006
Capstone Presentation
Junguk Hur
22
Overview
Differential Location Analysis - Summary
Background
Location
Analysis
Differential
Binding
TF response
classification
Comparison of TF binding
between rich medium and stress cond.
Represented by PiK
Distribution of PiK
Normal and Skewed
Chi-Square Tests
70% TF-cond pairs differential distribution
Classification of TF responses
UP, DOWN, and BOTH
05-26-2006
Capstone Presentation
Junguk Hur
23
How differential binding?
Overview
Background
Location
Analysis
Differential
Binding
TF Response
Classification
Comp. with
Microarray
1. Changes in the TF expression
2. Different TFBS signature for different category
3. Interaction with other TFs expression
4. Modifications in TFs (protein level)
5. Changes in physical structures
(epigenetic features)
6. Other unknown reasons
05-26-2006
Capstone Presentation
Junguk Hur
24
Overview
Comparison with Microarray - I
Background
Location
Analysis
Differential
Binding
TF Response
Classification
Comp. with
Microarray
ChIP-Chip
vs
Microarray data
---------------------expression levels
of transcription
factors (TFs).
---------------------H2O2 Low
(oxidation)
---------------------TF-Cond pairs : 28
Skewed : 23
Normal : 5
05-26-2006
Capstone Presentation
Junguk Hur
25
Overview
Comparison with Microarray - II
Background
Location
Analysis
Differential
Binding
TF Response
Classification
Comp. with
Microarray
SM
(Sulfometuron methyl)
Amino acid starvation
-------------------------ChIP-Chip
0.2 g/ml final
2 hours
-------------------------Microarray
0.2 g/ml 15 minutes
5.0 g/ml 4 hours
-------------------------TF-Cond pairs : 34
Skewed : 21
Normal : 13
05-26-2006
Capstone Presentation
Junguk Hur
26
Overview
Comparison with Microarray - III
Background
H2O2 Oxidation stress
Location
Analysis
Differential
Binding
TF Response
Classification
Comp. with
Microarray
SM (Sulfometuron methyl) Amino acid starvation
05-26-2006
Capstone Presentation
Junguk Hur
27
Overview
Comparison with Microarray - IV
Background
Location
Analysis
Differential
Binding
TF Response
Classification
Comp. with
Microarray
50~60% of differentially binding
TFs have NO significant
expression change
Changes of gene regulation
(differential binding) cannot be
fully revealed by Microarray
05-26-2006
Capstone Presentation
Junguk Hur
28
How differential binding?
Overview
Background
Location
Analysis
Differential
Binding
TF Response
Classification
Comp. with
Microarray
1. Changes in the TF expression
2. Different TFBS signature for different category
3. Interaction with other TFs expression
4. Modifications in TFs (protein level)
5. Changes in physical structures
(epigenetic features)
6. Other unknown reasons
05-26-2006
Capstone Presentation
Junguk Hur
29
Conclusion
Overview
Background
Location
Analysis
Categorization of Transcription Factors
Differential
Binding
(TFs) depending on the responses to stress
TF Response
Classification
Comp. with
Microarray
Conclusion /
Future Work
No significant expression changes in
about 50% of the tested TFs in response to
environmental changes.
Importance of integrating differential
binding of TFs with gene expression to get
the bigger picture
05-26-2006
Capstone Presentation
Junguk Hur
30
Future Work
Overview
Background
Location
Analysis
Differential
Binding
TF Response
Classification
Comp. with
Microarray
Diff. Binding
Motif Anal.
Investigating the biological mechanism of the
differentially binding of TFs
1. Target gene of expression level
2. Post-translational modification
3. Protein-protein interaction
Comprehensive differential analysis
TF-TF Corr.
Diff. Expression
Conclusion /
Future Work
Integration
Diff. Location
Diff. Proteomics (open)
Diff. Protein-Protein Int. (open)
05-26-2006
Capstone Presentation
Junguk Hur
31
References
Overview
Background
Location
Analysis
1.
Differential
Binding
2.
TF Response
Classification
Comp. with
Microarray
Diff. Binding
Motif Anal.
3.
4.
TF-TF Corr.
Conclusion /
Future Work
5.
Harbison, C.T., et al., Transcriptional regulatory code of a
eukaryotic genome. Nature, 2004. 431(7004): p. 99-104. .
Shapira, M., E. Segal, and D. Botstein, Disruption of yeast
forkhead-associated cell cycle transcription by oxidative
stress. Mol Biol Cell, 2004. 15(12): p. 5659-69.
Gasch, A.P., et al., Genomic expression programs in the
response of yeast cells to environmental changes. Mol Biol
Cell, 2000. 11(12): p. 4241-57.
Causton, H.C., et al., Remodeling of yeast genome
expression in response to environmental changes. Mol Biol
Cell, 2001. 12(2): p. 323-37.
Jia, M.H., et al., Global expression profiling of yeast
treated with an inhibitor of amino acid biosynthesis,
sulfometuron methyl. Physiol Genomics, 2000. 3(2): p. 8392.
05-26-2006
Capstone Presentation
Junguk Hur
32
Acknowledgements
Dr. Haixu Tang
Dr. Sun Kim
Dr. Mehmet Dalkilic
Dr. Predrag Radivojac
Seung-hee Bae
Sourav Roy
Capstone class 2006
School of informatics
Center for Genomics and Bioinformatics
Dr. Zhixiong Xue at DuPont
My Family
05-26-2006
Capstone Presentation
Junguk Hur
33
05-26-2006
Capstone Presentation
Junguk Hur
34
05-26-2006
Capstone Presentation
Junguk Hur
35
05-26-2006
Capstone Presentation
Junguk Hur
36
How differential binding?
Overview
Background
Location
Analysis
Differential
Binding
TF Response
Classification
Comp. with
Microarray
Diff. Binding
Motif Anal.
1. Changes in the TF expression
2. Different TFBS signature for different category
3. Interaction with other TFs expression
4. Modifications in TFs (protein level)
5. Changes in physical structures
(epigenetic features)
6. Other unknown reasons
05-26-2006
Capstone Presentation
Junguk Hur
37
Different TFBS signatures?
Overview
Background
Location
Analysis
Category “BOTH” has
similar numbers of
Up-differentially bound
and Down-differentially
bound regions.
Differential
Binding
TF Response
Classification
Comp. with
Microarray
Diff. Binding
Motif Anal.
Is there any difference
between these TFBSs in
different groups?
05-26-2006
Capstone Presentation
Junguk Hur
38
Different TFBS signatures?
Overview
Background
Location
Analysis
BOTH
Differential
Binding
TF Response
Classification
Up seqs
Comp. with
Microarray
Diff. Binding
Motif Anal.
Step1. Collect putative TFBS
A. Frankel’s motifs (102TFs)
B. Gary Stormo’s “Patser”
Down seqs
Step1
TFBSs
TFBSs
Step2. Create profiles from
collected TFBS seqs
Step2
Profile
Profile
Step3
05-26-2006
Step3. Profile comparison
A. “MatCompare” program
by Michale Zhang
Capstone Presentation
Junguk Hur
39
Different TFBS signatures?
Overview
Background
High-Threshold
Location
Analysis
Differential
Binding
TF Response
Classification
Comp. with
Microarray
Diff. Binding
Motif Anal.
05-26-2006
Capstone Presentation
Junguk Hur
40
Different TFBS signatures?
Overview
Background
Low-Threshold
Location
Analysis
Differential
Binding
TF Response
Classification
Comp. with
Microarray
Diff. Binding
Motif Anal.
05-26-2006
Capstone Presentation
Junguk Hur
41
How differential binding?
Overview
Background
Location
Analysis
Differential
Binding
TF Response
Classification
Comp. with
Microarray
Diff. Binding
Motif Anal.
TF-TF Corr.
1. Changes in the TF expression
2. Different TFBS signature for different category
3. Association with other TFs
4. Modifications in TFs (protein level)
5. Changes in physical structures
(epigenetic features)
6. Other unknown reasons
05-26-2006
Capstone Presentation
Junguk Hur
42
TF-TF Correlation
Overview
Background
Location
Analysis
Differential
Binding
TF Response
Classification
Comp. with
Microarray
Diff. Binding
Motif Anal.
TF-TF Corr.
TF
TF
05-26-2006
Capstone Presentation
Junguk Hur
43