Meet the scientists!

Biologists can now produce lots of measurements, and I work to try to develop models to help to interpret the data and understand the biological processes. For example, biologists are now able to use technology such as a gene expression microarray to measure the expression level of every gene in an organism.

There are a number of model organisms that many people study, such as mouse, fly, c. elegans (a tiny worm) and Arabidopsis (a weed). I’ve done some work that aims to learn a gene regulatory network for Arabidopsis. This uses lots of different microarrays, done in lots of different experimental conditions, which measure how the plant’s gene expression changes over time, or in response to an environmental change, such as different light conditions, or stresses such as salt, drought, cold, or pests. There is now data from over 4000 different experiments available online. For this large set, we can use a computer to search for patterns in the data. However, there are over 25,000 genes in Arabidopsis. We are able to find co-expressed genes – the ones that show a similar pattern across lots of experiments, and hence are more likey to be involved in similar biological processes. Also, we can use all this data to begin to learn a gene regulatory network for small parts of the network around genes of interest. We use a statistical algorithm to find the most likely network that generated the observed data. The network hypotheses can then be tested in the laboratory, in order to confirm true regulatory interactions and increase our understanding of how things work.

I tend to arrive at work at 9am. I spend most of my day working at a computer. This involves reading articles from Computer Science or Machine Learning Journals (specifically the part of Computing that deals with using a computer to learn models from data), Bioinformatics Journals or Biology Journals. I spend some time thinking about how things work, what data exists, and what kind of questions we can ask. I will write some code to manipulate data, and perform some analysis. A lot of these activities can take a long time, and I’ll probably work on several things at once.

On a practical note, I generally eat lunch (a couple of sandwiches) with the rest of the BioSystems group at about 12:30 and discuss what we’re working on, or talk about what’s in the news. I go to several talks a week, some about genes, others about statistics, maths, application areas. These vary from student presentations, or discussion groups, to presentations from world-leading scientists.

I leave work at about 5:30, and go running for an hour most nights. I’m now looking forward to when the clocks change and I’ll be able to run off-road in the evenings.

In September, along with five other scientists from the UK working with GeneChips, Computers and Crops, I am going to run a summerschool in Beijing with about 20-30 chinese students and 5-6 UK PhD students, to teach them about what we do. This will cover information about the technology i.e. microarrays (gene chips) used to measure gene expression levels, about data analysis and interpretation, and about gene networks. Specifically we hope to investigate ginseng, and analyse the different expression patterns in different tissue types. I will put the money towards this, as it promises to be a useful experience for students from different cultures and backgrounds to meet and engage with each other and learn more about microarray data analysis.