Translating the Cell - PowerPoint PPT Presentation

Loading...

PPT – Translating the Cell PowerPoint presentation | free to download - id: 56fada-YzYzM



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Translating the Cell

Description:

Title: Slide 1 Author: Rachel Created Date: 1/28/2008 6:59:36 PM Document presentation format: On-screen Show Company: BWH Other titles: Arial Times New Roman ... – PowerPoint PPT presentation

Number of Views:33
Avg rating:3.0/5.0
Slides: 43
Provided by: Rac9106
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Translating the Cell


1
Translating the Cells Instruction Manual A
Biophysicists Approach to Understanding Gene
Regulation
  • Rachel Patton McCord
  • Bulyk Lab
  • Harvard University Biophysics Program
  • 3/20/08

2
  • Knobloch lives?
  • What are characteristics of life?
  • Response to environment
  • Take in nutrients and produce waste
  • Reproduction
  • .

3
Biological Signal Processing
oxygen
ethanol
4
Biological Signal Processing
Inputs
Outputs
protein
Transcription Factor
mRNA
Nucleus
5
Regulation of Gene Expression
  • Transcription Factor (TF) recognizes DNA bases
    (ACGT)
  • Promotes gene expression transcription of mRNA

RNA Polymerase
RNA
Sequence-Specific TFs
(output)
6
Organisms
  • Ideal understand gene regulation in human
  • Problems Large genome size, diverse cell types,
    likely complicated gene regulation rules
  • Begin with model system single celled organism
    Saccharomyces cerevisiae (yeast)

7
Goals
  • Find DNA sequences bound by TFs
  • Predict how TFs function in the cell
  • Look for biophysical links between TF structure
    and function
  • Use quantitative approaches to maintain a
    physically realistic view of biology.

8
TF-DNA Sequence Recognition
Protein Binding Microarray (PBM) Technology
dsDNA
Fluorophore labeled antibody
Microarray slide
Mukherjee, Berger, et al., Nature Genetics
(2004), 361331-1339.
9
TF-DNA Sequence Recognition
Protein Binding Microarray (PBM) Technology
Detector
Laser
(488 nm)
Mukherjee, Berger, et al., Nature Genetics
(2004), 361331-1339.
10
Universal Array Design
  • Interested in sequences of 8-10 bases

410 1,000,000 total 10-mers
410 1,000,000 total 10-mers 410 / 27 40,000
total spots
24 nt fixed sequence
36 nt variable sequence
5
3
CTATCTACACACAACTATGCGGTCGCCATGGAAATGGTCTGTGTTCCGTT
GTCCGTGCTG
CTATCTACACA
TATCTACACAC
27 10-mers per spot
ATCTACACACA
TCTACACACAA
Berger, Philippakis et al., Nature Biotechnology
(2006), 241429-1435.
Philippakis, Qureshi et al., RECOMB (2007).
11
Universal Array Design
  • Use an idea from cryptography
  • de Bruijn sequence contains all sequence
    variants of length k in the shortest sequence
    possible

de Bruijn sequence
All possible 3-mers
AAA AAC AAG AAT ACA ACC ACG
ACT AGA AGC AGG AGT ATA ATC ATG
ATT CAA CAC CAG CAT CCA CCC CCG
CCT CGA CGC CGG CGT CTA CTC CTG
CTT GAA GAC GAG GAT GCA GCC GCG
GCT GGA GGC GGG GGT GTA GTC GTG
GTT TAA TAC TAG TAT TCA TCC TCG
TCT TGA TGC TGG TGT TTA TTC TTG
TTT
Test sequence (36 bp)
TCGATTGCGTGACAGGGTAAAACAAGACCCTGACCATGGCAGTGT
TCGATTGCGTGACAGGGTAGTCCGGGTTCTTTGCGCTCACTATAC
Length 43 64 bp
Fixed sequence (24 bp)
Anthony Philippakis, Mike Berger
12
Deriving Binding Strength at each Sequence
  • Every 8mer is represented 16 times
  • Take median over intensities of all spots
    containing this 8mer

Example CATGGAAA
CCGTCAGCAGTCATGGAAAGCTGGTAGAAGTTCTGGGTCTGTGTTCCGTT
GTCCGTGCTG TTATACCATGGAAAGACAAACGTAGCATGTTGGAGTGTC
TGTGTTCCGTTGTCCGTGCTG CCATGGAAATGTGTCCCTAAGGGTGGTA
ACAAAATAGTCTGTGTTCCGTTGTCCGTGCTG CACTACGCAAGTGCGGT
GCATGGAAAGGGTTCTGGAGTCTGTGTTCCGTTGTCCGTGCTG ATCTCA
TGGAAAAGACTCATAACGATCAACAGTCGGGTCTGTGTTCCGTTGTCCGT
GCTG ACAACAGAGCACCGATGGCATGGAAACTTGCGTAGAGTCTGTGTT
CCGTTGTCCGTGCTG GTGGAGAAAGGGGTCAAACATGGAAACGCATCGA
CAGTCTGTGTTCCGTTGTCCGTGCTG GCCCGGGATCCCATCCATGGAAA
ATGTCGCTTACATGTCTGTGTTCCGTTGTCCGTGCTG CAGAAGTGTCCT
ACGTAACATCCACATGGAAAGTACGTCTGTGTTCCGTTGTCCGTGCTG G
TTGCATACACGCATGGAAATAACAATCGAACTCCAGTCTGTGTTCCGTTG
TCCGTGCTG TCATGTGCTGGGCTTGATTCAGCATGGAAAACCAGTGTCT
GTGTTCCGTTGTCCGTGCTG TATTCTTCTCTTCATGGAAACAGTAAAAA
ATCGGACGTCTGTGTTCCGTTGTCCGTGCTG CTATCTACACACAACTAT
GCGGTCGCCATGGAAATGGTCTGTGTTCCGTTGTCCGTGCTG CCTGGGG
ACATGGAAAAATGAAGTCACCCATGGTGCGTCTGTGTTCCGTTGTCCGTG
CTG ATCATCCTTACATTACATGGAAATCGTGTGCCAATAGTCTGTGTTC
CGTTGTCCGTGCTG AAGGCCCATGGAAACCACGTCATATTCACAACTAA
CGTCTGTGTTCCGTTGTCCGTGCTG
13
Deriving Binding Strength at each Sequence
8-mer
Rev. Comp.
Median Signal
GTCACGTG CACGCGAC 108178 GCACGTGC GCACGTGC
95854 CACGTGCC GGCACGTG 89203 GCACGTGA TCACGTGC
74295 TCACGTGA TCACGTGA 69377 ACACGTGA TCACGTG
T 68733 ATCACGTG CACGTGAT 58874 CACGTGTA TACAC
GTG 58656 CCACGTGA TCACGTGG
47900 ACACGTGG CCACGTGT 47240 CACGTGAG CTCACGTG
42887 AGCACGTG CACGTGCT 41755 ACACGTGC GCACGTG
T 36764 CACGTGTC GACACGTG 36463 ACCACGTG CACGT
GGT 36380 CACGTGCG CGCACGTG
35515 CACGTGCA TGCACGTG 32370 AACACGTG CACGTGTT
28948 CCACGTGC GCACGTGG 22983 CACGTGGC GCCACGT
G 19315 ... ... ...
ka
kd
ka
TF DNA TF-DNA
kd
14
Goals
  • Find DNA sequences bound by TFs
  • PBMs
  • Predict how TFs function in the cell
  • Look for biophysical links between TF structure
    and function
  • Use quantitative approaches to maintain a
    physically realistic view of biology.

15
Predicting TF Cellular Functions
  • Use known/measurable inputs and outputs

Gene expression
Heat shock Gene Deletion
mRNA
16
Gene Expression Data
  • 1327 Publicly Available Microarray Datasets

Condition 1
mRNA
Condition 2
17
Predicting Cellular Functions of Components
  • Basic model/assumptions
  • TF binding near genes causes change in expression
  • Similar TF binding probability similar
    expression active regulation

Expression data
PBM data
TF1
Gene 1
TF1
Gene 2
TF1
Gene 3
TF1
Gene 4
Gene 5
18
Physically Realistic Binding Probability
  • Simple (and often used) view

Promoter region is BOUND Gene is ON
Cbf1
GGCACGTGGCTGCATGAGCGGAGTCACGTGGGAAAATACAACAGTCACCC
ACGTGCCGTGCACCGACGTACTCGCCTCAGTGCACCCTTTTATGTTGTCA
GTGGGTGCAC
Gene
Promoter region is NOT BOUND Gene is OFF
GGCACGTGGCTGCATGAGCGGAGGCTCGCGGGAAAATACAACAGTCACCC
ACGTGCCGTGCACCGACGTACTCGCCTCCGTGCGCCCTTTTATGTTGTCA
GTGGGTGCAC
Gene
19
Physically Realistic Binding Probability
  • Physical reality
  • Energy landscape of potential TF binding
  • TF occupancy probability Integration of binding
    potential across sequence near gene
  • Dictates likelihood of recruiting RNA polymerase
    and thus level of mRNA transcription

Cbf1
GGCACGTGGCTGCATGAGCGGAGTCACGTGGGAAAATACAACAGTCACCC
ACGTGCCGTGCACCGACGTACTCGCCTCAGTGCACCCTTTTATGTTGTCA
GTGGGTGCAC
Gene
20
Physically Realistic Binding Probability
  • Physical reality
  • Energy landscape of potential binding
  • Sum median intensity data across all possible
    8-mers in sequence near gene

Cbf1
GGCACGTGGCTGCATGAGCGGAGTCACGTGGGAAAATACAACAGTCACCC
ACGTGCCGTGCACCGACGTACTCGCCTCAGTGCACCCTTTTATGTTGTCA
GTGGGTGCAC
Gene
Intensity 117651
Intensity 215352
GGCACGTGGCTGCATGAGCGGAGTCACGTGGGAAAATACAACAGTCACCC
ACGTGCCGTGCACCGACGTACTCGCCTCAGTGCACCCTTTTATGTTGTCA
GTGGGTGCAC
Gene
21
Goals of New Analysis Method
  • Combine binding probability with expression data
    to predict TF function and condition specific
    binding site usage

Target Gene 1 2 3 4 5 6
Gene expression
PBM data
Condition A
Condition B
Condition C
Condition D
TF Function
22
Goals of New Analysis Method
  • Consider all data rather than drawing arbitrary
    cutoffs
  • Low affinity binding as well as minor expression
    changes may be biologically relevant
  • Tanay, 2006 Foat et al., 2006

Binding probability
?
23
CRACR
  • Combination Rank-order Analysis of
    Condition-specific Regulation

24
Basics of CRACR Approach
  • Order genes by expression in condition of
    interest
  • Assign ranks based on PBM-derived binding
    probability for TF

25
Basics of Analysis Approach
  • Select
  • similarly expressed foreground genes
  • background set

PBM p-value rank
background
2 3 6 9 1 8
5 10 4 7 11
Most
Most induced
repressed
YGR087C
YAR003W
YGR043C
YAR044W
YER130C
YPL054W
YAR014C
YGR088W
YAR029W
YAR018W
YAL003C
26
Basics of Analysis Approach
  • Slide window along ordered expression
  • Calculate an area statistic for enrichment of PBM
    targets within each window vs. background

PBM p-value rank
2 3 6 9 1 8
5 10 4 7 11
Most
Most induced
repressed
YGR087C
YAR003W
YGR043C
YAR044W
YER130C
YAR014C
YPL054W
YGR088W
YAR029W
YAR018W
YAL003C
27
Predicting TF Function
  • Plot area statistic (ranges -0.5 to 0.5) at each
    window
  • Determine condition significance by permutation
    test-derived threshold (gray line p lt 0.001)

metabolism switch
metabolism enzyme
Glucose added Mig1 targets repressed
Glucose
area statistic
Mig1
induced-----------------repressed
mRNA
gt8.0 5.0 3.4 2.3 1.5
0 -1.5 -2.3 -3.4 -5 lt-8
Expression fold change
28
Predicting TF Function
  • Determine which individual genes are repressed by
    Mig1

Group of genes repressed by Mig1
Glucose added Mig1 targets repressed
Mig1
YHR005C
Mig1
area statistic
YER130C
Mig1
YBL054W
induced-----------------repressed
gt8.0 5.0 3.4 2.3 1.5
0 -1.5 -2.3 -3.4 -5 lt-8
Expression fold change
29
Prediction of General TF Function
  • Find all (of 1327) expression conditions where a
    TF is predicted to be active
  • Look for enrichment of general biological
    functions in this set

Selected Mcm1 significant conditions
Conditions for which there is significant enrichment of PBM targets  Effect
Cell Cycle Expression in response to Clb2p (set 1, 40 min) induced
Expression during the cell cycle (alpha factor arrest and release)(16) induced
Expression during the cell cycle (cdc15 arrest and release)(8) induced
Expression during the cell Cycle (cdc28)(7) induced
Expression in response to 50 nM alpha-factor 120 min induced
Expression in ckb2 deletion mutant induced
Expression in dig1, dig2 deletion mutant induced
Expression in swi6 (haploid) deletion mutant induced
Expression in tec1 (haploid) deletion mutant induced
Expression in yel044w deletion mutant induced
Expression in sir2 deletion mutant repressed
Expression in snf2 mutant cells in minimal medium repressed
Expression in response to 50 nM alpha-factor in bni1mutant 60 min repressed
30
Prediction of General TF Function
  • Find all (of 1327) expression conditions where a
    TF is predicted to be active
  • Look for enrichment of general biological
    functions in this set

Selected Mcm1 significant conditions
Conditions for which there is significant enrichment of PBM targets  Effect
Cell Cycle Expression in response to Clb2p (set 1, 40 min) induced
Expression during the cell cycle (alpha factor arrest and release)(16) induced
Expression during the cell cycle (cdc15 arrest and release)(8) induced
Expression during the cell Cycle (cdc28)(7) induced
Expression in response to 50 nM alpha-factor 120 min induced
Expression in ckb2 deletion mutant induced
Expression in dig1, dig2 deletion mutant induced
Expression in swi6 (haploid) deletion mutant induced
Expression in tec1 (haploid) deletion mutant induced
Expression in yel044w deletion mutant induced
Expression in sir2 deletion mutant repressed
Expression in snf2 mutant cells in minimal medium repressed
Expression in response to 50 nM alpha-factor in bni1mutant 60 min repressed
31
Prediction of General TF Function
  • Find all (of 1327) expression conditions where a
    TF is predicted to be active
  • Look for enrichment of general biological
    functions in this set
  • Prediction Mcm1 involved in cell cycle and mating

Selected Mcm1 significant conditions
alpha factor
alpha cell
a cell
32
Prediction of TF function
  • After PBM experiments, CRACR has been used to
    predict functions of 90 yeast TFs (paper in
    process)

33
Binding Site Affinity Effects
TF concentration low
TF
High affinity
TF concentration medium
Gene 1
TF concentration high
Binding affinity
TF
Medium affinity
Gene 2
TF
Low affinity
Gene 3
ka
ka
kd
TF DNA TF-DNA
kd
34
Demonstrating Effects of Binding site affinity
  • Low vs. high affinity binding sites may have
    different biological functions

Experimentally Validated
Expression after oxidative stress vs. Rap1
binding affinity
Highest binding affinityLowest
binding affinity
35
Goals
  • Find DNA sequences bound by TFs
  • PBMs
  • Predict how TFs function in the cell
  • CRACR
  • Look for biophysical links between TF structure
    and function
  • Use quantitative approaches to maintain a
    physically realistic view of biology.

36
Reasons for Different Functions TF structure?
  • Goal Consider biophysical TF structure instead
    of cartoon TF blob

tup1
cyc8
Mig1
37
TF Structure and Function
  • Are certain TFs structurally suited for certain
    types of biological processes?
  • Case Study

CST6 (bZIP)
Lower Information Content Motif
Regulatory hub many target genes
cell fate, cell cycle
GAL4 (Zn2Cys6)
Higher Information Content Motif
More specific, fewer target genes
metabolism of specific nutrients
38
Goals
  • Find DNA sequences bound by TFs
  • PBMs
  • Predict how TFs function in the cell
  • CRACR
  • Look for biophysical links between TF structure
    and function
  • Use quantitative approaches to maintain a
    physically realistic view of biology.

39
Future Directions
  • Completion of functional predictions and study of
    yeast gene regulation
  • Toward predictive model in humans
  • Experiments for understanding gene regulation
    rules

40
Acknowledgements
  • Martha Bulyk
  • Mike Berger
  • Anthony Philippakis
  • Cong Zhu
  • Kelsey Byers
  • Trevor Siggers
  • Vicky Zhou
  • Cherelle Walls
  • Jason Warner
  • Jaime Chapoy
  • Other Bulyk Lab Members

NSF graduate research fellowship NIH/NHGRI R01
41
GO CATS!!
42
Advantages and Challenges of Interdisciplinary
Work
  • Insight gained by quantitative reasoning in
    biology, combining of different perspectives
  • Physicists and mathematicians choose projects in
    biology that are fun, but not necessarily
    important
  • Important not to get caught up in what counts
    as true biology or true physics
About PowerShow.com