Katal

1
Network biology in cancer
Prof. Peter Csermely LINK-Group,Semmelweis
University,Budapest, Hungary
www.linkgroup.hu csermelynet_at_gmail.com
2
Traditional view
cause
effect
(Paul Ehrlichs magic bullet)
3
Recently changed view
100 causes
100 effects
4
Networks may help!
major causes
major effects
5
Advantages of the network approach

• Networks have general properties
• small-worldness
• hubs (scale-free degree distribution)
• nested hierarchy
• stabilization by weak links

Watts Strogatz, 1998
Karinthy, 1929
Barabasi Albert, 1999
Csermely, 2004 2009

• Generality of network properties offers
• judgment of importance
• innovation-transfer across different layers of
  complexity

6
Influential nodes in different systemsexample
to break conceptual barriers
Aging is an early warning signal of a critical
transition
death

• Prevention nodes with less predictable behaviour
• omnivores, top-predators
• market gurus
• stem cells

Farkas et al., Science Signaling 4pt3
7
Adaptation of complex systems
homeostasis
stress
Norbert Wiener
Ludwig von Bertalanffy
Conrad Waddington
cybernetics
homeorhesis
8
A possible adaptation mechanism
9
Plasticity and rigidity two key, but ill-defined
concepts
stability
stability
complexity
complexity
robustness
robustness
emergent property
emergent property
degeneracy
Plasticity functional structural
Rigidity functional structural
learning
learning
memory
memory
evolution
evolution
evolvability
evolvability
canalization
scientific revolution
exploitation (focus)
exploration (diversify)
creativity
aging
aging
creativity
10
Plasticity and rigidity two key, but ill-defined
concepts
11
Definition of functional plasticity and rigidity
large number of responses
small number of responses
12
Functional plasticity and rigidity and system
stability
plastic systems smooth state space
rigid systems rough state space
simple systems
smooth perturbation (not necessarily small)
13
Plasticity-rigidity cycles form a general
adaptation mechanism
Plasticity
Rigidity
alternating changes of plasticity- and
rigidity-dominance allow the recalibration of the
system to find the maximal structural stability
in a changed environment
14
Properties of plastic and rigid systems
extremely plastic
structurally stable, robust
extremely rigid


possibility of adaptation


effect of adaptation
Gáspár Csermely, Brief. Funct. Genom.
11443 Gyurkó et al. Curr. Prot. Pept. Sci. 15171
15
Example 1 Molecular mechanismsof protein
structure optimization
Hsp60 chaperone
unfolded substrate (plastic)
folded substrate (rigid)
chaperone cycle
substrate expansion (rigid)
substrate release (plastic)
Hsp60 iterative annealing pull/release of
folding protein
Todd et al, PNAS 934030 Csermely BioEssays
21959 Lin Rye, Mol. Cell 1623
16
Example 2 cell differentiation cancer attractors
progenitor
Sui Huang
Ingemar Ernberg
differentiated cells
Stuart Kauffman
Huang, Ernberg, Kauffman, Semin. Cell Developm.
Biol. 20869
17
Example 3 cell differentiation
more rigid
differentiated cells
rigid
progenitor cells
plastic
Rajapakse et al., PNAS 10817257
gene expression correlation networks chromatin
networks
18
Example 4 disease progression
Scientific Reports 2342 813
phosgene inhalation-induced lung injury, chronic
hepatitis B/C, liver cancer
19
Example 5 cancer stem cells
Csermely et al., Seminars in Cancer Biology doi
10.1016/j.semcancer.2013.12.004
20
Network-independent mechanisms of
plasticity-rigidy cycles
1. noise reaching hidden attractors coloured
noise, node-plasticity 2. medium-effects water,
chaperones membrane-fluidity, volume
transmission as neuromodulation, money
21
Network-dependent mechanisms of
plasticity-rigidy cycles
soft spots creative nodes, prions (Q/N-rich
proteins), chaperones
rigidity seeds rigidity promoting nodes

• extended, fuzzy core
• fuzzy modules
• no hierarchy
• source-dominated

• small, dense core
• disjunct, dense modules
• strong hierarchy
• sink-dominated

Csermely et al., Seminars in Cancer Biology doi
10.1016/j.semcancer.2013.12.004
22
Topologicalphase transitionsplastic ? rigid
networks with diminished resources
star network
scale-free network
random graph
complexity
subgraphs
stress
resources
Derényi et al., Physica A 334583
23
Yeast stress induces module condensation of the
interactome

• Stressed yeast cell
• nodes belong to less modules
• modules have less contacts
• more condensed modules
• more separated modules

• yeast protein-protein interaction
• network 5223 nodes, 44314 links
• several other conditions
• stress 15 min 37C heat shock
• other 4 stresses
• link-weight changes mRNA
• expression level changes

Mihalik Csermely PLoS Comput. Biol. 7e1002187
24
Drug design strategiesfor plastic and rigid cells
e.g. antibiotics
e.g. rapamycin
Csermely et al, Pharmacol Therap 138 333-408
25
Central hit network-influence cancer
cancer stem cells
most test systems are in this stage
most patients are in this stage
Gyurkó et al, Seminars in Cancer Biology
23262-269
26
network entropy low high
János Hódsági, MSc thesis
27
Network entropy increases than decreases in
cancer propagation
plastic
adenoma
colon
rigid
carcinoma
János Hódsági MSc thesis
28
Drug design strategiesfor plastic cells
e.g. antibiotics
e.g. rapamycin
Csermely et al, Pharmacol Therap 138 333-408
29
3 novel network centralities reveal influential
nodes
perturbation centrality
(www.Turbine.linkgroup.hu) community
centrality (www.modules.linkgroup.hu) game
centrality (www.NetworGame.linkgroup.hu)
PLoS ONE 5e12528 Bioinformatics 282202 Science
Signaling 4pt3 PLoS ONE 8e67159 PLoS ONE
8e78059
30
Bridges are key nodes of social regulation
hispanic
old
union leaders strike
sociogram leaders work
BC
BC
BC
Hawk-dove game (PD game same) Start
all-cooperation strike Strike-breaker
defects BC-s are the best strike-breakers
young
Farkas et al., Science Signaling 4pt3 Simko
Csermely PLoS ONE 8 e67159 www.linkgroup.hu/Netw
orGame.php Michaels strike network Michael,
Forest Prod. J. 4741
31
3 novel network centralities reveal influential
nodes
perturbation centrality
(www.Turbine.linkgroup.hu) community
centrality (www.modules.linkgroup.hu) game
centrality (www.NetworGame.linkgroup.hu)
PLoS ONE 5e12528 Bioinformatics 282202 Science
Signaling 4pt3 PLoS ONE 8e67159 PLoS ONE
8e78059
32
ModuLand method family module centres bridges
community landscape
influence zones of all nodes/links
community centrality a measure of the
influence of all other nodes
communities as landscape hills
network hierachy
Szalay-Beko et al. Bioinformatics 282202
extensive overlaps centre of modules bridges
available as Cytoscape plug-in
Kovacs et al, PLoS ONE 5e12528 www.modules.linkgr
oup.hu network of network scientists Newman PRE
74036104
33
Drug design strategiesfor rigid cells
e.g. antibiotics
e.g. rapamycin
Csermely et al, Pharmacol Therap 138 333-408
34
Network-influence Allo-network drugs
hit of intra- cellular paths

• Examples BRAF inhibition
• restoring MEK inhibition
• rapamycin effects on
• mTOR complexes

• atomic resolution interactome
• of allosteric protein complexes
• identification of allosteric paths

Nussinov et al, Trends Pharmacol Sci 32686
35
Network influence Multi-target drugs
Csermely et al, Trends Pharmacol Sci 26178
36
3 novel network centralities reveal influential
nodes
perturbation centrality
(www.Turbine.linkgroup.hu community
centrality (www.modules.linkgroup.hu) game
centrality (www.NetworGame.linkgroup.hu
PLoS ONE 5e12528 Bioinformatics 282202 Science
Signaling 4pt3 PLoS ONE 8e67159 PLoS ONE
8e78059
37
Turbine general network dynamics tool
any real networks can be added,
modified normalizes the input network any
perturbation types (communicating vessel model,
multiple, repeated, etc.) any models of
dissipation, teaching and aging Matlab compatible
www.Turbine.linkgroup.hu
Szalay Csermely, Science Signaling 4pt3 PLoS
ONE 8e78059
38
Attractors of T-LGL network using
TurbineAttractor
apoptosis
proliferation
39
Multi-drug design with TurbineDesigner
T-LGL survival signaling network leukemia
specific edges Starting state IL7-activation
target-state all black TurbineDesigner
solution to reach target state
apoptosis
starting state
Inactive protein
Inactive protein
Activated protein
Activated protein
Network Zhang R, Shah MV, Yang J, Nyland SB,
Liu X, Yun JK, Albert R, Loughran TP Jr. (2008)
Network model of survival signaling in large
granular lymphocyte leukemia. PNAS 105 1630813.
40
Take-home messages
41
Acknowledgment the LINK-Group the associated
talent-pool
A core of 8 people a multidisciplinary group of
34 people with a background of 100 members
and a HU/EU-talent support network