Transcript Madan-Neuro-Oct2008-final - MRC Laboratory of Molecular

Dissecting the dynamics of transcriptional
regulatory networks
M. Madan Babu
Group leader
MRC Laboratory of Molecular Biology
Cambridge, UK
Overview of research
Evolution of biological systems
Evolution of networks within and across genomes
Uncovering a distributed architecture in networks
Evolution of transcriptional networks
Nature Genetics (2004)
J Mol Biol (2006a)
PNAS (2008)
Structure, function and regulation of biological systems
Structure and dynamics of biological networks
Science (in press)
Nature (2004)
Methods to investigate network dynamics
Principles of network dynamics
Submitted
Data integration and function prediction
Discovery of transcription factors in pathogens
Discovery of novel DNA binding proteins
C
Evolution of a global regulatory hubs
H
H
C
Nuc. Acids. Res (2005)
J Bacteriol (2008)
Nuc. Acids. Res (2006)
Outline
Structure of the transcriptional regulatory network
Global dynamics (network dynamics) of regulatory networks
Local dynamics (node dynamics) of regulatory networks
Networks in Biology
Network
Protein Interaction
Metabolic
Transcriptional
Nodes
Proteins
Metabolites
Transcription factor
Target genes
Links
Physical Interaction
Enzymatic
conversion
Transcriptional
Interaction
Protein-Protein
Protein-Metabolite
Protein-DNA
Interaction
A
A
A
A
B
B
B
B
Organization of the transcriptional regulatory network
analogy to the organization of the protein structures
Basic unit
Transcriptional interaction
Local structure - Motifs
Patterns of interconnections
Global structure - Scale free network
All interactions in a cell
Local structure
Secondary structure
Global structure
Class/Fold
Transcription
factor
Target gene
Basic unit
Peptide link between amino acids
Transcriptional networks are made up of motifs
Function
ArgR
AraBAD
- Responds to persistent signal
- Filters noise
ArgF
AraC
Multiple input
Motif
TrpR
Crp
ArgD
“Patterns of
interconnections
that recur at different
parts and
with specific
information
processing task”
Single input
Motif
Feed Forward
Motif
ArgE
Network Motif
- Co-ordinates expression
- Enforces order in expression
- Quicker response
AroM
TyrR
AroL
- Integrates different signals
- Quicker response
Shen-Orr et. al. Nature Genetics (2002) & Lee et. al. Science (2002)
Transcriptional networks are “scale-free”
Scale-free structure
Presence of few
nodes with many links and many
nodes with few links
N (k) a
1
g
k
“Scale-free” structure provides robustness to the system
Albert & Barabasi, Rev Mod Phys (2002)
Summary I – Structure of transcriptional networks
Transcriptional networks are made up of motifs that have
specific information processing task
Transcriptional networks have a scale-free structure which confers
robustnessto such systems, with hubs assuming importance
Madan Babu M, Luscombe N et. al
Current Opinion in Structural Biology (2004)
Outline
Structure of the transcriptional regulatory network
Global dynamics (network dynamics) of regulatory networks
Local dynamics (node dynamics) of regulatory networks
Global dynamics of the regulatory networks
Static regulatory
network in yeast
Cell cycle
Sporulation
~ 7000 interactions involving
3000 genes and 142 TFs
Stress
Across all cellular conditions
How does the local structure change in different cellular conditions?
How does the global structure change in different conditions?
Regulatory program specific transcriptional networks
Stress
Binary
Processes
Regulatory
programs
involved in
survival
Diauxic shift
Multi-step
Processes
DNA damage
Sporulation
Cell cycle
Regulatory
programs
involved in
development
Pre-sporulation
Sporulation
E L M
Germination
G0 G1
S G2 M
Temporal dynamics of local structure
Multi-step regulatory
programs (development)
Binary regulatory
programs (survival)
fast acting
&
direct
fast acting
&
direct
slow acting
&
indirect
Network motifs that allow for efficient execution of regulatory steps
are preferentially used in different regulatory programs
Temporal dynamics of global structure (hubs)
Condition
specific hubs
Hubs regulate other hubs to
trigger cellular events
Each regulatory
program is triggered
by specific hubs
Permanent hubs
Active across all
regulatory programs
This suggests a dynamic structure which transfers
‘power’ between hubs to trigger distinct regulatory
programs (developmental & survival)
multi-stage conditions
Cell cycle
Sporulation
• fewer target genes per TF
• longer path lengths
• more inter-regulation
between TFs
Fidelity in response
binary conditions
Diauxic shift
DNA damage
Stress
• more target genes per TF
• shorter path lengths
• less inter-regulation
between TFs
Quick response
Sub-networks re-wire both their local and global
structure to respond to cellular conditions efficiently
Luscombe N, Madan Babu M et. al
Nature (2004)
Summary II – Temporal dynamics of network structure
Network motifs are preferentially used by different regulatory
programs. This allows for efficient response to different
cellular conditions
Different regulatory proteins become global regulatory hubs
under various cellular conditions. This allows for transition between
regulatory programs by switching on condition specific hubs
Luscombe N, Madan Babu M et. al
Nature (2004)
Outline
Structure of the transcriptional regulatory network
Global dynamics (network dynamics) of regulatory networks
Local dynamics (node dynamics) of regulatory networks
Dynamics in transcription and translation
Each node in the network represents several entities (gene, mRNA, and protein)
and events (transcription, translation, degradation, etc) that are compressed in
both space and time
Higher-order organization of transcriptional regulatory networks
1
10
4
4
1
9
3
5
5
2
3
6
6
2
10
7
8
Inherent higher-order
Flat structure
organization
9
7
8
Hierarchical structure
Hierarchical organization of the yeast regulatory network
Top
layer
Transcription
Factor
Topological
Sort
Core
layer
Bottom
layer
Target gene
Bottom (59)
AZF1
HAL9
ZAP1
AFT1*
DAL81
GLN3
AFT2*
ARG81
FHL1*◄ FKH1
GTS1
GZF3
ASH1
FKH2*
HAP1
CBF1*
GAL4
HAP4
CIN5*
GAT1
HCM1*
HSF1*◄
MSN2*
INO2
MSN4*
INO4
NDT80
IXR1
PHD1*
REB1*◄ RIM101
SUT1
SWI4*
YAP1
YAP5*
ROX1
SWI5
YAP6*
RPN4
TEC1*
YAP7*
SKO1
TOS4*
YHP1
LEU3
PLM2*
SMP1
TOS8*
YOX1*
MGA1*
PUT3
SOK2*
TYE7
UME6*
CAT8
HMRA2 SPT23
SIP4
STP2
BAS1
HMLALPHA1
MATALPHA1
HMS1 FZF1
HAC1
PHO4
STP1
PDR1
RGT1
SUM1
UGA3
YAP3
YDR026C
MET28 MIG3
PDR3
STB4
STB5
YDR049W
MET4◄
ACA1
CRZ1
CST6
GCR1◄
PDR8
CAD1
ECM22
MAC1
MBP1*
Level 6
ABF1*◄
MAL33
ACE2
CUP9
GAT3
ADR1
DAL80
GCN4*
Level 5
Level 4
Level 3
Level 2
Level 1
HMLALPHA2 HMS2
MIG1
MIG2
RAP1*◄
STE12*
XBP1
RTG3
HMRA1
RME1
IME2
SFL1
SRD1
YRR1
MOT3
PPR1
PDC2◄
ARR1
DAT1
RTG1
CUP2
LYS14
MSN1
RDS1
MAL13
MSS11
RGM1
EDS1
ARO80
FLO8*
PIP2
CHA4
MET31
MCM1◄
ARG80
DAL82
NRG1* MET32
OAF1
PHO2
SKN7*
Level 7
OTU1 RPH1
Core (64)
Top (25)
Target Genes
RDR1
RLM1
THI2
UPC2
YBL054W
YDR266C
YJL206C
YKR064W
YRM1
SFP1
WAR1
URC2
USV1
STP4
YER130C YER184C YML081W YPR196W
* Regulatory hubs
◄ Essential genes
Hierarchical organization of regulatory proteins
Regulatory proteins are hierarchically organized
into three basic layers
Top layer (29)
Core layer (57)
Bottom layer (58)
Target Genes
Do TFs in the different hierarchical levels have
distinct dynamic properties?
Datasets characterizing dynamics of transcription and translation
Transcript abundance
Transcript half-life
Protein abundance
Protein half-life
(Holstege et al., 1998)
(Wang et al., 2002)
(Yang et al., 2003)
(Ghaemmaghami et al, 2003)
(Belle et al, 2006 )
Transcript half-lives were
determined by obtaining
transcript levels over several
minutes after inhibiting
transcription. This was done
using the temperature
sensitive RNA polymerase
rpb1-1 mutant S. cerevisiae
strain.
Estimates of the
endogenous protein
expression levels
during log-phase were
obtained by TAPtagging every yeast
protein for S.
cerevisiae.
Transcript abundances
for yeast grown in YPD
(S. cerevisiae) and
Edinburgh minimal
medium (S. pombe) were
determined by using an
Affymetrix high density
oligonucleotide array.
Transcriptional dynamics
Protein half-lives were
determined by first
inhibiting protein
synthesis via the addition
of cyclohexamide and by
monitoring the abundance
of each TAP-tagged
protein in the yeast
genome as a function of
time.
Translational dynamics
Regulation of regulatory proteins within the hierarchical framework
Regulation of transcript
abundance or degradation
does not appear to be a major
control mechanism by which
the steady state levels of TFs
are controlled
Regulation of protein
abundance and degradation
appears to be a major control
mechanism by which the
steady state levels of TFs are
controlled
post-transcriptional regulation plays an important role in ensuring
the availability of right amounts of each TF within the cell
Noise in protein levels in a population of cells
Noise in a population of cells can be beneficial where phenotypic diversity could be
advantageous but detrimental if homogeneity and fidelity in cellular behaviour is required
No. of copies
of Protein i
5
6
12
1
4
3
3
8
1
3
9
1
19
2
7
10
0
8
No of copies
Noisy
No of copies
Frequency
Almost
no noise
Frequency
Frequency
Population of cells
Very Noisy
No of copies
Noise levels of regulatory proteins
Top layer
Core layer
Bottom layer
Target Genes
Less noisy
More noisy
Regulatory proteins in the top layer are more noisy than the ones in
the core or bottom layer
Implication
Individual 1
Underlying
network
Individual 2
Individual 3
Individual 4
Differential utilization of the same underlying network
by different individuals in a population of cells
Noise in TF expression may permit differential utilization of the same
underlying regulatory network in different individuals of a population
Noise in expression of a master regulator of sporulation in yeast
Gene network controlling
sporulation
High variability in the expression of
top-level TFs in a population of cells
may confer a selective advantage as
this permits at least some members in a
population to respond quickly to
changing conditions
Nachman et al Cell (2008)
Summary III – local dynamics of regulatory networks
Our results suggest that the core- and bottom-level TFs are more tightly
regulated at the post-transcriptional level rather than at
the transcriptional level itself
Our findings suggest that the interplay between the inherent hierarchy of the
network and the dynamics of the TFs permits differential utilization of the
same underlying network in distinct members of a population
Conclusions
Global
dynamics
Local
dynamics
Different regulatory programs preferentially
use different network motifs
Top, core and bottom level TFs display
distinct patterns in protein abundances and
half-lives
Condition specific hubs allow for transition
between regulatory programs
Top level TFs are more noisy than the core
or bottom level TFs
May permit robust response to changing
conditions by efficiently re-wiring the
active regulatory network
May provide flexibility in response in a
population of cells, making the organism
more adaptable to changing conditions
Regulatory networks are dynamic within the timescale of an organism and
the interplay between the global and local dynamics makes the network
both robust and adaptable to changing conditions
Acknowledgements
Nick Luscombe
EBI, Hinxton, UK
Raja Jothi
S Balaji
NCBI, NIH, USA
Mark Gerstein
Haiyuan Yu
Mike Snyder
Yale University, USA
Sarah Teichmann
MRC-LMB, UK
Arthur Wuster
Joerg Gsponer
MRC-LMB, UK
Joshua Grochow
Chicago University, USA
L Aravind
Teresa Przytycka
NCBI, NIH, USA
MRC - Laboratory of Molecular Biology, UK
National Institutes of Health, USA
Schlumberger and Darwin College, Cambridge, UK
Topological Sort: An approach to infer hierarchical organization in networks
1. Identify strongly connected
components (SCCs)
2. Obtain Directed Acyclic Graph
(DAG) by collapsing the SCCs
3. Construct the transpose of DAG
(reverse the direction of edges)
10
10
10
4
9
5
3
1
1
1
9
9
3,4,
5,6
3,4,
5,6
6
2
2
7
11
11
8
2
7,8
11
7,8
4a. Run iterative leaf-removal algorithm 4b. Run iterative leaf-removal algorithm
Construct
Identify
Obtain
Directed
strongly
the transpose
Acyclic
connected
Graph
of DAG
Invert the layers
components
(DAG)
(reverse
bythe
collapsing
direction
(SCCs)
the
of hierarchy
edges)
SCCs
Construct
the
final hierarchy
Run iterative
Combine
to
deduce
leaf-removal
the
algorithm
Layer 4
1
Layer 3
3,4,
5,6
Layer 2
Layer 1
Layer 4
9
11
Layer 3
2
7,8
2
10
Layer 2
3,4,
5,6
10
7,8
9
Layer 1
1
11
6. Combine to
7. Construct the final hierarchy
1 11
Layer 4
1
11
Layer 3
3
Layer 2
2
4
5
6
Layer 3
3,4,
5,6
Layer 2
2
10
hierarchy
Layer 4
5. Invert the levels
Layer 4
1
11
Layer 3
3,4,
5,6
10
Layer 2
2
7,8
Layer 1
9
10
7,8
Layer 1
7
8
9
Layer 1
9
Methods to infer hierarchical organization in networks
Food web network
Leaf-removal algorithm
BFS-level algorithm
?
predator
Topological sort
6
prey
5
4
4
4
3
3
3
2
2
2
1
1
1
Applicable on cyclic networks
Consistent ordering
Scalable
Allows for ambiguity (e.g. Mouse)
2-4
Yeast transcriptional regulatory network
Transcription
Factor (TF)
156 TFs (regulatory proteins)
4,403 TGs (proteins being regulated)
12,702 links (regulatory interactions)
Target gene (TG)
In-degree (#incoming connections) = 2.9
Out-degree (#outgoing connections) = 81.4
On average, each gene is regulated by ~3 TFs
On average, each TF regulates ~80 genes
Topological properties of regulatory proteins within the hierarchical framework
Top
Top 85.4
Top
1
Core
Core
All
Top
Core
Bottom
Bottom
Unclassified
Core 145.8
y = 43.706x-1.4426
R2 = 0.87
Frequency
Bottom
Bottom
1771
28.2
20.7
Top
3639
43.9
Core
p < 10-4
1152
Bottom 0.0
y = 1.8904x-0.7092
R2 = 0.57
Target Genes
0.1
y = 1E-08x3 - 8E-06x2 + 0.0015x + 0.0122
0
80
160
R2 = 0.47
Avg # TGs (out-degree)
y = 145.13x-1.8078
R2 = 0.92
0.01
0
100
0
2000
Total number of TGs
y = 8.7209x-1.0941
R2 = 0.92
200
TF out-degree
300
4000
400
0
25
50
% TFs that are hubs
Genetic and Evolutionary aspects of regulatory proteins within
the hierarchical framework
Top
70.6
Top
13.8
Top
p < 0.0644
Core
p < 0.041
5.3
Core
65.3
Core
p < 3x10-3
Bottom
Bottom 3.4
56.9
Bottom
p < 10-4
Target Genes
0
5
10
15
% TFs that are essential
40
60
80
% Conservation
Temporal dynamics of global structure
Condition specific networks are all scale-free
Cell cycle
Sporulation
Internal & multi-step
regulatory programs
(Development)
Diauxic shift
DNA damage
Stress
External & binary regulatory programs
(Survival)
Do different proteins become hubs under different conditions?
Is it the same protein that acts as a regulatory hub across multiple conditions?
Higher-order organization in yeast regulatory network
Top layer
Transcription
Factor (TF)
Core layer
Bottom layer
Target gene (TG)
Top Layer
(29)
Core Layer
(57)
Target Genes
ABF1
ARG80
ARO80
CHA4
DAT1
HAL9
INO2
IXR1
ARR1
MAC1
AZF1
HSF1
MAL33
MBP1
MCM1
MET31
MET32
NDT80
NRG1
OAF1
PIP2
PDC2
PHO2
RPH1
RTG1
SKN7
UME6
OTU1
ZAP1
ACE2
ADR1
AFT1
AFT2
ARG81
ASH1
CBF1
CIN5
PPR1
CUP9
DAL80
DAL81
FHL1
FKH1
FKH2
GAL4
GAT1
RDR1
GAT3
GCN4
GLN3
GTS1
GZF3
HAP1
HAP4
HCM1
RLM1
HMLALPHA2
INO4
LEU3
MIG1
MIG2
MSN2
MSN4
PHD1
STP4
PLM2
PUT3
RAP1
REB1
RIM101
ROX1
RPN4
SKO1
THI2
SMP1
SOK2
STE12
SUT1
SWI4
SWI5
TEC1
TOS4
UPC2
TOS8
TYE7
XBP1
YAP1
YAP5
YAP6
YAP7
YHP1
YBL054W
ACA1
BAS1
CAD1
CAT8
CRZ1
CST1
CUP2
ECM22
YJL206C
FZF1
GCR1
HAC1
HMALPHA1
HMRA1
HMRA2
HMS1
HMS2
YKR064W
LYS14
MAL13
MATALPHA1
MET28
MET4
MGA1
MIG3
MOT3
YRM1
MSN1
MSS11
PDR1
PDR3
PDR8
PHO4
RDS1
RGM1
RGT1
RME1
RTG3
SFL1
SFP1
SIP4
SPT23
SRD1
STB4
STB5
STP1
STP2
SUM1
UGA3
WAR1
YAP3
YBR033W
YDR026C
YDR049W
YDR520C
YER130C
YER184C
YML081W
YPL230W
YPR196W
YRR1
FLO8
YOX1
Bottom Layer
(58)
YDR266C
Unclassified (12)
A
BFS-level algorithm on Yeast transcription network
Level 4
ARG81
MAC1
NDT80
Level 3
ADR1
GLN3
OAF1
AFT2*
GTS1
PDC2◄
ARG80
HMS1
PIP2
ASH1
INO4
RPN4
AZF1
IXR1
SPT23
CAT8
LEU3
TOS8*
CUP9
MAL33
TYE7
DAL81
MCM1◄
YAP1
DAL82
MIG1
FKH2*
MIG2
Level 2
ABF1*◄
FHL1*◄
HAL9
MBP1*
PHD1*
ROX1
STP1
XBP1
ACE2
FKH1
HAP1
MET31
PHO2
RPH1
STP2
YAP3
AFT1*
FLO8*
HAP4
MET32
PHO4
RTG1
SUM1
YAP5*
ARO80
FZF1
HCM1*
MET4◄
PLM2*
RTG3
SUT1
YAP6*
ARR1
GAL4
HMLALPHA2
MGA1*
PUT3
SIP4
SWI4*
YAP7*
CBF1*
GAT1
HMRA1
MSN2*
RAP1*◄
SKN7*
SWI5
YDR026C
CHA4
GAT3
HMRA2
MSN4*
REB1*◄
SKO1
TEC1*
YHP1
CIN5*
GCN4*
HMS2
NRG1*
RGT1
SMP1
TOS4*
YOX1*
DAL80
GZF3
HSF1*◄
OTU1
RIM101
SOK2*
UGA3
ZAP1
DAT1
HAC1
INO2
PDR1
RME1
STE12*
UME6*
Level 1
ACA1
IME1
PDR8
STB5
YER130C
BAS1
LYS14
PPR1
STP4
YER184C
CAD1
MAL13
RDR1
THI2
YJL206C
CRZ1
MATALPHA1
RDS1
UPC2
YKR064W
CST6
MET28
RGM1
URC2
YML081W
CUP2
MIG3
RLM1
USV1
YPR196W
ECM22
MOT3
SFL1
WAR1
YRM1
EDS1
MSN1
SFP1
YBL054W
YRR1
GCR1◄
MSS11
SRD1
YDR049W
HMLALPHA1
PDR3
STB4
YDR266C
TFs with zero in-degree
* Regulatory hubs
◄ Essential genes
An instance of inaccurate hierarchy
as inferred by the BFS-level algorithm
B
Level 3
OAF1
Level 2
UME6
Level 2
GAT1
Strongly
Connected
Component
NDT80
Level 4
Level 4
ARG81
AFT2
Level 3
Level 3
CUP9
AFT1
Level 2
Level 2
YAP6
RDS1
ACA1
IME1
XBP1
YER130C
RGM1
STB4
YML081W
Level 1
“Scale-free” networks exhibit robustness
Robustness – The ability of complex systems to maintain their function even
when the structure of the system changes significantly
Tolerant to random removal of nodes (mutations)
Vulnerable to targeted attack of hubs (mutations) – Drug targets?
Hubs are crucial components in such networks
Temporal dynamics of global structure (hubs)
Each regulatory
program is triggered
by specific hubs
Presence of permanent
hubs across all
regulatory programs
Permanent hubs
Existence of condition
specific hubs
Condition specific hubs
Connectivity profile: No of regulated genes
CC
SP
DS
DD
SR
TF
250
45
20
30
15
Swi6
Hubs regulate other hubs to trigger cellular events
This suggests a dynamic structure which transfers ‘power’ between hubs
to trigger distinct regulatory programs (developmental & survival)