Transcript Folie 1 - Tresch Group

Modeling Combinatorial
Intervention Effects in
Transcription Networks
(The Sound of One-Hand Clapping)
Achim Tresch
Computational Biology
Gene Center Munich
The Question
If two hands clap and there is a sound;
what is the sound of one hand?
(Japanese Kōan)
Kōan
A paradoxical anecdote or riddle, used in Zen Buddhism to demonstrate
the inadequacy of logical reasoning and to provoke enlightenment.
Synthetic Genetic Interactions
How to define “Interaction“
mathematically?
ΔB
GrowthYB of single
manipulation of B
GrowthYA of single
manipulation of A
Synthetic
Genetic
Array
ΔA
ΔA
ΔB
Growth YAB
of double
manipulation
of A and B
modified after Collins, Krogan et al., Nature 2007
Synthetic Genetic Interactions
How to define “Interaction“
mathematically?
Phenotype
Measurement YA
of single perturbation
ΔB
Phenotype
Measurement YB
of single perturbation
Phenotype
Measurement YAB
of double perturbation
ΔA
The interaction score SAB is a function
of the two single perturbations and the
combined perturbation,
SAB = SAB (YA ,YB ,YAB )
ΔA
ΔB
Synthetic Genetic Interactions
Common Interaction Scores
Define an expected phenotype of the double perturbation as a
function f(YA ,YB ) of the single perturbation phenotypes YA and Yb.
The interaction score SAB is then the deviation from the expected
phenotype
SAB = YAB - f(YA ,YB )
Common choices for f :
f = min(YA ,YB )
(v. Liebig´s minimum rule for plant growth)
f = YA ·YB
(chemical equilibrium a + b ↔ ab , [a][b] = [ab])
f = YA + YB
(log version of YA ·YB )
f = log2[(2YA - 1)(2YB - 1) + 1] (essentially the same as YA + YB )
Interaction Scores
are not very reliable
Results crucially
depend on f
Mani, Roth et al., PNAS 2007
Synthetic Genetic Interactions
Breakthrough: Combine a set of weak predictors to create a strong predictor
(guilt by association = correlation of interaction scores)
Pan, Boeke et al., Cell 2006
Collins, Krogan et al., Nature 2007
Cartoon by Van de Peppel et al, Mol. Cell 2005
Synthetic Genetic Interactions
Take home message: Two components are likely to interact (physically)
whenever they have the same interaction partners
Costanzo M, Myers CL, Andrews BJ, Boone C, et al.: Science 2010
Screening for TF interactions
If two hands clap and there is a sound;
what is the sound of one hand?
ΔA
One manipulation
High dimensional
readout
Genetic interactions from one perturbation
Step 1: Construct a transcription factor - target graph
a) From ChIP binding experiments
Harbison, Fraenkel, Young et al.
Nature 2004
MacIsaac, Fraenkel et al.
BMC Bioinformatics 2006
b) From protein binding arrays, followed by PWM-based predictions
Ansari et al., Nature Methods 2010
Berger, Bulyk et al., Nature Biotech 2006
Genetic interactions from one perturbation
Step 1: Construct a transcription factor - target graph
Intersection size of target sets of TF1 and TF2 can be used alone to assess
TF cooperativity.
(Beyer, Ideker et al., PlOS Comp. Biol 2006)
Genetic interactions from one perturbation
Step 2: Combine TF-target information and expression data
~2.000
target genes
118 transcription
factors
Established Methods for the
detection of univariate TF activity :
GSEA
(Subramanian, Tamayo PNAS 2005)
Globaltest
(Goemann, Bioinformatics 2004)
MGSEA
(Bauer, Gagneur, Nucl. Acids Res. 2010)
and many more …
Graph obtained from MacIsaac
et al. (BMC Bioinformatics 2006)
Common Idea: A TF is active if
its set of target genes shows
significantly altered expression.
To quantify this, various tests
are constructed.
Genetic interactions from one perturbation
Step 3: Given TF1 and TF2, group genes into 4 interaction classes
Binding sites
TF 1
TF 2
Synthesis rates during salt stress
gene 1
gene 2
TF1
gene 3
gene 4
TF 1 is
active
TF 2 is
active
TF2
TF1
TF 1+2
active
TF2
time
Antagonistic interaction of TF 1+2
Genetic interactions from one perturbation
Step 3: Given TF1 and TF2, group genes into 4 interaction classes
Binding sites
TF 1
TF 2
Synthesis rates during salt stress
gene 1
gene 2
TF 1 is
inactive
gene 3
TF 2 is
inactive
gene 4
TF 1+2
active
time
Synergistic interaction of TF1+2
Genetic interactions from one perturbation
Step 4: Use these 4 groups to define an interaction score
For any pair of transcription factors T1 and T2, we perform a
logistic regression.
 P( g is induced ) 
 ~  0
log 
 P( g is not induced ) 
 1  Ind ( g is a target of TF1)
(for all genes g)
  2  Ind ( g is a target of TF2)
 12  Ind ( g is a target of TF1 and TF2)
Our interaction score for the pair (T1,T2) is then β12.
Genetic interactions from one perturbation
Step 4: Use these 4 groups to define an interaction score
. . . ~  0  1  Ind (TF1  g )   2  Ind (TF2  g )  12  Ind (TF1 , TF2  g )
Example:
Binding sites
TF 1
TF 2
0  0
gene 1
~ 0
 0  1  0
TF 1 is
active
0   2  0
TF 2 is
active
gene 2
~  0  1
gene 3
~ 0 
gene 4
~  0  1 
2
 2  12
 0  1   2  12  0
12  0  (0  1 )  ( 0  1 )  0
time
Antagonistic interaction
TF 1+2
active
Application: Osmotic stress in yeast
Use the guilt by association trick to construct an interaction matrix for all
transcription factors using only a two group microarray comparison!
Inclusion
criterion:
only TFs
with >70
targets
„One hand clapping“
Miller, Tresch,
Cramer et al.,
Mol. Syst.
Biol. 2010,
in revision
Application: Osmotic stress in yeast
Validation with
BioGRID database:
Among 84 TFs
under consideration
(with enough targets),
3486 potential
interactions
Exist. Only 97
interactions are
recorded.
Application: Osmotic stress in yeast
Validation with
BioGRID database:
Single interactions scores
don‘t work well
Profile correlations
do work
Genetic interactions from one intervention
One hand clapping can be applied to:
Microarray data, Pol II ChIP data, nascent RNA data
Application to a similar dataset
leads to similar results:
3 stress responses:
osmotic stress NaCl,
osmotic stress KCl,
heat shock
(Mitchell, Pilpel at al. Nature 2009):
Acknowledgements
Gene Center Munich:
Patrick Cramer
Dietmar Martin
Björn Schwalb
Sebastian Dümcke
20
My Answer
Two hands clap and there is a sound;
what is the sound of one hand?
It is similar for transcription factors that interact.
Systems Buddhism
Zen Biology