Genetic Switches

One of my research interests is genetic switches. A simple genetic switch is shown below.

The 'rod' in the figure represents a piece of DNA, possibly part of the bacterial chromosome. The light blue and pink rectangles are the DNA sequences encoding the light blue and pink proteins, shown as circles. The coloured arrows indicate that proteins are produced from the DNA. In fact, this is a complex process. The green gene encodes Green Fluorescent Protein (GFP) which can be seen under the microscope and is used to follow what is happening to the bacteria. The DNA also contains 'regulatory sequences', shown in the figure as the dark blue and brown rectangles. The binding of protein molecules to the DNA at these sequences controls the production of proteins from the DNA. In this network, the pink protein binds to the DNA at the dark blue regulatory sequence. This prevents the light blue protein from being produced. Likewise, the light blue protein binds to the brown regulatory sequence, preventing production of the pink protein. GFP production is also prevented if the blue protein binds to the brown sequence. The net result is that this network is bistable: it has two alternative stable steady states, one in which a large amount of pink protein, but little blue protein, is present (here GFP is abundant and the bacteria are green), and the other where the blue protein, but not the pink, is abundant (here GFP is not abundant and the bacteria are not green). Because all biological systems are subject to random fluctuations, the network is expected to flip occasionally between these two stable states. One of the subjects of my work is predicting this flipping rate, using computer simulations, and comparing the results of the simulations to experimental measurements.

A switch like the one shown in the figure has been been made experimentally and shown to be bistable in the bacterium E. coli [T. S. Gardner, C. R. Cantor & J. J. Collins, Nature 403, 339 (2000)]. Lucas Black and I are currently measuring its switching rate for different growth rates of the bacteria.

A very interesting "real" switch is the lysis-lysogeny switch of the bacteriophage lambda. This switch determines whether the virus lambda will multiply inside its host bacterium, killing the bacterium and releasing many progeny viruses (this is called lysis), or whether it will insert itself into the DNA of the bacterium and remain dormant there for many bacterial generations (this is called lysogeny). Many questions remain to be answered about this switch (which seems to be extremely stable to fluctuations), and computer simulations may be a good way to do this. Marco Morelli, Pieter Rein ten Wolde and I have just published a paper in which we use stochastic simulations to simulate the flipping of this switch. Our simulations show that a recently discovered DNA looping interaction can stabilise the switch against intracellular destabilising factors such as nonspecific binding of transcription factors to genomic DNA.

Rare event simulation methods

In order to study the flipping of the genetic switch shown in the figure, we needed to develop a simulation method for rare events. These are processes that happen rapidly, yet very infrequently (such as the flipping of the switch). Such events are notoriously difficult to simulate, because for conventional simulations, the vast majority of the computational effort is spent on the uninteresting 'waiting time' between events and few, if any, events are observed in a typical simulation run. Our technique, which we call 'Forward Flux Sampling', or FFS, avoids this problem by focusing only on the event itself. The aim is to calculate the rate at which the rare event happens, and to obtain a collection of simulation trajectories corresponding to the event, which can be analysed to obtain information on the mechanism. These trajectories are members of the 'Transition Path Ensemble'. The FFS method is illustrated below.

FFS can be applied to many different types of rare events, not just the flipping of genetic switches. What is particularly useful about it is that the system to be simulated does not need to be in equilibrium (which is not the case for most rare event simulation methods). This means that we can study, for example, polymer translocation through pores when a pulling force is applied, or how crystals nucleate when there is a shear flow in the surrounding liquid. So far, FFS has been used for nucleation of NaCl crystals, the diamond-graphite transition of carbon, folding of a lattice model of a protein, droplet coalescence and heterogeneous nucleation in a 2D Ising model. I am currently working on improvements to the method, to make it more efficient and user-friendly, and I am always looking for interesting new applications.

Microbial Ecosystems

I collaborate with Dr. Andrew Free of the School of Biological Sciences in Edinburgh, on a joint experimental/simulation project to understand the development and ultimately evolution of self-sustaining microbial ecosystems. We have been making and analysing Winogradsky Columns - which are samples of pond sediment and water, exposed to light for a period of weeks or months. These columns develop into self-sustaining ecosystems. We are analysing using modern molecular techniques the species composition of these columns and we are developing a computational model to predict and understand their development. This project is relevant to many important questions in ecology and evolution. The Winogradsky column is also a beautiful example of a non-equilibrium "steady state". This work has been funded by a grant from the Leverhulme Trust. The picture shows the "Winogradsky team" and some onlookers collecting samples.

Microbial response to hydrostatic and osmotic pressure

I collaborate with Dr. Bruce Ward of the School of Biological Sciences in Edinburgh, on an experimental project which is carried out by a PhD student, S. Lucas Black. We are interested in how bacteria respond to hydrostatic and osmotic pressure. In particular, we investigate the growth of the deep sea bacterium Photobacterium profundum as a function of hydrostatic and osmotic pressure. Together with Dr. Hugh Vass, Lucas has developed a microscope pressure cell for imaging bacteria under high pressure. He has also developed a method for growing bacteria in 96 well plates under pressure and is currently using this to screen mutant libraries. This work is assisted by close contacts with Prof. Doug Bartlett of the Scripps Institute of Oceanography and with Dr. Gail Ferguson of Aberdeen University.