Protein Networks
Systems and Synthetic Biology
498A
Copyright © 2008: Sauro
1
Yeast Interaction Network
Jeong H, Mason SP, Barabási AL and Oltvai ZN (2001) Lethality
and centrality in protein networks. Nature 411:41-42.
Copyright © 2008: Sauro
2
Electronic Network
Copyright © 2008: Sauro
3
Protein Network
Walter KOLCH, Biochemical Journal (2000) Volume 351, 289-305
Copyright © 2008: Sauro
4
Protein Network
Copyright © 2008: Sauro
Kitano: CellDesigner
5
Attempts at Standardization
Kohn Diagrams
Explicit
Heuristic
Combinatorial
Depicting combinatorial complexity with the
molecular interaction map notation.
MSB, 2(51), 2006. Kohn et al.
Copyright © 2008: Sauro
6
Attempts at Standardization
Example of Explicit Map
Copyright © 2008: Sauro
7
Attempts at Standardization
Example of Heuristic Map
Copyright © 2008: Sauro
8
SBGN: Systems Biology Graphical Notation
Glycolysis
Copyright © 2008: Sauro
Simple Protein Network
9
SBGN: Systems Biology Graphical Notation
More Complex Protein Network
Copyright © 2008: Sauro
10
Electrical Engineering
Copyright © 2008: Sauro
11
Electrical Engineering
1954, Popular Mechanics
12
Copyright © 2008: Sauro
Electrical Engineering
http://web.cecs.pdx.edu/~harry/Relay/
Copyright © 2008: Sauro
13
Elemental Processes in Protein Networks
Copyright © 2008: Sauro
14
Cell Division
Copyright © 2008: Sauro
15
Cell Cycle Control Network
Modelling the controls of the eukaryotic cell cycle. B. Novák and J.J. Tyson.
Biochemical Society Transactions (2003) 31, (1526–1529)
Copyright © 2008: Sauro
16
Protein Networks
It has been estimated from experimental studies that in E. coli, 79
proteins can be phosphorylated on serine, threonine and tyrosine side
groups whereas in Yeast, 4000 phosphorylation events involving 1,325
different proteins have been recorded.
By comparison, E. coli has 171 transcription factors.
In higher eukaryotic cells, particularly human, around 2% of the protein
coding part of the genome is devoted to encoding protein kinases, with
perhaps 10% of the coding region dedicated to proteins involved in
signaling networks. It has also been suggested that possibly as much as
30% of all cellular proteins in yeast and human can be phosphorylated
Copyright © 2008: Sauro
17
Protein Cycles
Up to 30% of the human protein-coding
genes encode components of signaling
pathways, including transmembrane
proteins, guanine-nucleotide binding
proteins (G proteins), kinases,
phosphatases and proteases.
The identification of 518 putative
protein kinase genes and 130 protein
phosphatases suggests that
reversible protein phosphorylation is
a central regulatory element of most
cellular functions.
18
Abundance of Kinases
Species
# of putative kinases
Saccharomyces cerevisiae
121
Drosophila melanogaster
319
C. elegans
437
Arabidopsis thal
1049
Human
518
Data from http://www.kinexus.ca
19
The Simple Cascade
v
2
v1
20
Response Characteristic
?
21
Linear Response
• A simple example of protein
dynamics: protein synthesis
and degradation.
22
Hyperbolic Response
Assume linear kinetics
23
Hyperbolic Response
24
Sigmoidal Response
Assume saturable kinetics
25
Sigmoidal Response
Assume saturable kinetics
26
Sigmoidal Response
M1 = R/(R+RP)
M2 = RP/(R+RP)
v1
v1
27
Sigmoidal Response
M1 = R/(R+RP)
M2 = RP/(R+RP)
v1
v1
28
Fundamental Properties
Ultrasensitivity
E1
E2
X
Kms = 0.5
29
Fundamental Properties
Ultrasensitivity
E1
E2
X
Kms = 0.1
30
Fundamental Properties
Ultrasensitivity
E1
E2
X
Kms = 0.02
31
Fundamental Properties
Ultrasensitivity
v1, v2
Km = 1, Vmax = 1
Vmax
v1
v2
E1
32
Fundamental Properties
Ultrasensitivity
v1, v2
Km = 0.03, Vmax = 1
Vmax
v1
v2
E1
33
Output
Collector Current
Device Analogs
Base Current
Input
Input
34
Digital Circuits
In ultrasensitive mode, cascades can be
used to build Boolean circuits.
NOT Gate
A
B
35
Ring Oscillator
36
NAND Gate
A B
A
B
C
C
37
Memory Units
Basic flip-flop
R = reset
S = set
Q = output
38
Memory Units
Clocked RS flip-flop
R = reset
S = set
C = clock
Q = output
39
Counters
Clock input
Binary Counter
0
0
0
0
1
0
0
0
0
1
0
0
1
1
0
0
0
0
1
0
1
0
1
0
0
1
1
0
1
1
1
0
Clock RS flip-flop
etc
40
Arithmetic
Half Adder (No carry input)
A
B
Sum
Carry
0
0
0
0
0
1
1
0
1
0
1
0
1
1
0
1
41
Sigmoidal Response
Multiple Cycles
The response of the cascade is the
product of the individual cycle
sensitivities.
Cascades can therefore have considerable
gain values (Hill coefficients >> 10).
42
Multiple Phosphorylation Sites
p53 is a transcription
factor that is involved
in preventing cancer
formation.
Physiologically it is
involved in a variety
of stress signals,
cell cycle and
programmed
death functions.
It has multiple
modification sites – 21
phosphorylation sites..
http://www.biology.bnl.gov/cellbio/human_p53.html
43
Multiple Phosphorylation Sites
Assume linear kinetics
S3
44
Multiple Phosphorylation Sites
Assume linear kinetics
S3
45
Sigmoidal Response
Oscillators
Cell-signalling dynamics in time and space
Boris N. Kholodenko Nature Reviews
Molecular Cell Biology 7, 165-176 (March
2006) |
46
Sigmoidal Response
Bistable Switches
Cell-signalling dynamics in time and space
Boris N. Kholodenko Nature Reviews Molecular
Cell Biology 7, 165-176 (March 2006) |
47
Sigmoidal Response
Oscillators
Cell-signalling dynamics in time and space
Boris N. Kholodenko Nature Reviews Molecular
Cell Biology 7, 165-176 (March 2006) |
48
Sigmoidal Response
Oscillators
Cell-signalling dynamics in
time and space
Boris N. Kholodenko
Nature Reviews Molecular
Cell Biology 7, 165-176
(March 2006) |
49
Amplifiers
Provided the feedback is below the
threshold to cause oscillations, feedback
systems can behave as robust amplifiers.
50
Cascades as Noise Filters
1
0.9
0.8
P2 / (P1 + P2)
0.7
30
0.6
0.5
20
0.4
0.3
10
0.2
0.1
0
0.5
1
1.5
2
V1 / V2
Cascades can act as
signal noise filters in
the most sensitive region
Magnitude
0
V1 = 1.7
0
V1 = 3.06
-10
V1 = 3.4
V1 = 3.74
-20
V1 = 5.1
-30
-40
-50
0.001
0.01
0.1
1
10
100
Frequency
Output
51
Why? Frequency Analysis
0
0
0.5
1
1.5
2
-5
Nr(dv/ds)L
-10
-15
Jacobian
-20
-25
-30
-35
V1/V2
1
0.9
0.8
P2 / (P1 + P2)
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0
0.5
1
1.5
2
V1 / V2
52