Transcript Slide 2

Regulatory networks
Biochemical noise
by Rosalind Allen
Slide 1: Cells with identical genes in identical environments can behave differently.
This can be explained in terms of biochemical noise. Cover image by Elowitz et al
from Science Volume 297 issue 5584. Reprinted with permission from AAAS. This figure
may be used for non-commercial, classroom purposes only. Any other uses require the prior written permission from AAAS.
q
A B 

C
Rate constant q is set to 1. Each simulation run is
repeated five times (results shown in different colours) .
Slide 2: Computer simulation results for reaction A + B →C, starting with different
numbers of A and B molecules.
Michael B Elowitz et al. 2002 Stochastic gene
expression in a single cell Science 298 1183.
Slide 3: Image demonstrating that genetically and environmentally identical cells
can show very different levels of gene expression. Cover image from
<i>Science</i> Volume 297 issue 5584. Reprinted with permission from AAAS. This
figure may be used for non-commercial, classroom purposes only. Any other uses require the prior written permission from
AAAS.
Intracellular
environment fluctuates:
this affects the two
genes in the same way
– the two colours are
correlated.
Randomness of
chemical reactions
involved in protein
production: this affects
the two genes
independently – the
two colours are
decorrelated.
Slide 4: Illustration showing the effects on the two reporter proteins of fluctuations
in the intracellular environment and of biochemical noise in transcription and
translation. From Elowitz <i>et al</i> 2002 Stochastic Gene Expression in a Single
Cell Science 297 1183. Reprinted with permission from AAAS. This figure may be used for
non-commercial, classroom purposes only. Any other uses require the prior written permission from AAAS.
Slide 5: The probability distribution p(N) (shown by the vertical bars) is changed by
the protein production and degradation reactions.
dp( N , t )
 kp( N  1)   ( N  1) p( N  1)  kp( N )  Np( N )
dt
N
Steady state: dp(N,t)/dt = 0
k
1  k  
  e
p( N ) 
N!   
hence
N 
k

N 
N
N

N2  N

k

2

1
N
Slide 6: The chemical master equation and probability distribution p(N) for the
simple “one-step” model of protein expression.
k

Average number of protein
molecules = 5
Average number of protein
molecules = 100
For the two-step model, we assume that on average an mRNA produces five
proteins, and we fix the transcription rate to get the same average number of
proteins as in the one-step model.
Slide 7: Protein number probability distributions for one-step and two-step models
of protein production.
“Venus” is a gene that encodes
a yellow fluorescent protein;
“tsr” encodes a peptide that
anchors the fluorescent protein
in the cell membrane.
Slide 8: An experimental system constructed by Yu et al. (2006) to visualise in real
time the production of a single protein molecule in a cell. From Yu et al. 2006
Probing gene expression in live cells, one protein molecule at a time, Science 311
1600. Reprinted with permission from AAAS. This figure may be used for non-commercial, classroom
purposes only. Any other uses require the prior written permission from AAAS.
The appearance of
individual protein
molecules was monitored
in growing cells over time.
Yellow dots show
individual protein
molecules bound to the
cell membrane.
The three plots show the
rate of protein production
(number of new proteins
per 3 minute time
interval) for three
different cell “lineages”,
which are illustrated on
the right.
Slide 9: Results showing the moment when individual protein molecules are produced in
growing bacterial cells. From Yu et al. 2006 Probing gene expression in live cells, one
protein molecule at a time, Science 311 1600. Reprinted with permission from AAAS. This figure may be
used for non-commercial, classroom purposes only. Any other uses require the prior written permission from AAAS.