Brian Marsden: From information to structure
Protein structures are powerful tools in the development of medical drugs, but they are not very accessible to non-specialists. Research informatics presents these structures more simply and interactively, and helps scientists make decisions. This will hopefully accelerate the development of new medicines.
Q: Can you tell us a little bit about your research interests?
Brian Marsden: I work for an organisation called the Structural Genomics Consortium or SGC, which is interested in trying to solve the structures of human proteins and making them available to the public. One of the key things we are particularly interested in is understanding how these protein structures inform or create disease. As part of the work associated with that, we generate a lot of data. My research is interested in how we capture that data and be able to ask questions to better understand how these mutations for example in proteins may cause disease.
Q: How do you use an informatics approach to support the work done at the SGC?
BM: We have a very multidisciplinary team at the SGC: we have people who are doing cloning of genes for example, creating proteins, solving the structures of proteins, even people who are doing medicinal chemistry making small molecules to interrogate those proteins. One of the things we have to do is to capture this information in a way that makes sense and is easy to mine. As a consequence of that, at the SGC we are entirely paperless and that means that everything that we do is captured in electronic format. This means that we have access to information from 10 years or more, no matter whether people are still with us or not. One of the other really important aspects of being able to have everything paperless, or electronically, is that we have the ability to ask questions or spot trends in the work that we are doing, finding out about what we are doing right, what we are doing wrong and spotting things that we would not have seen otherwise.
Q: How can visualisation tools help with drug discovery?
BM: At the SGC we solve many protein structures of human proteins. These make very pretty pictures but they are not particularly accessible to those people who are not structural biologists. It is really important that we find ways to make this information available to those who do not understand how to interpret these structures. Something we have pioneered at the SGC over that last 10 years is the ability to show these protein structures in a context that makes sense to the scientist. For example, we can show protein structures in a web browser, directly connected with a narrative, in an interactive form. This means that people like medicinal chemists for example can now look at these protein structures and understand the context of what that actually means, and help them make decisions about what sort of compounds they make.
Q: What are the most important lines of research that have emerged in the last 5-10 years?
BM: When we started at the SGC over 10 years ago, the sequencing of the human genome was a really big thing and we expected that would be a big game changer in terms of what we do. I would say that the revolution hasn't happened as quickly as we would have expected but these days it is possible to sequence a human genome within a day or so for less than a $1,000. That means that you can now ask questions like "what exactly is a single cell doing on its own in an organism?" We can now understand the interplay between how cells in the pancreas for example, or in the brain may be working in respect to each other. No two single cells are the same, so this means that what those cells are doing may be very different.
We can also ask bigger questions, we can do something called metagenomics - this is the study where we look at environmental effects that may be occurring in a bacterial colony. A classic example may be the bacterial colony that you have in your gut. It has been found that the flora you have in the gut may actually influence whether you have irritable bowel disease for example. By sequencing all these bacteria we can have a better understanding how that interplay may occur within a disease.
Q: Why does this line of work matter and why should we fund it?
BM: Drug discovery is an incredibly expensive thing to do, and there are no guarantees that a drug that you think you have is going to work out. Something we want to do at the SGC is to find ways to find the diamonds in the rough effectively. What are the right targets? What are the right genes? What are the right proteins that we should be working on to get new medicines quicker, more cheaply and in a more relevant way?
Q: How does your work fit into translational medicine within the department?
BM: Recently at the SGC we have been awarded a large strategic award by the Wellcome Trust which is focused on us developing something called target enabling packages. These build upon the work we have done at the SGC for many years: protein structures, how to make proteins, what small molecules we should use etc., and packaging them up together into something which people within academia and industry can go away and use to better understand whether the protein associated with those packages are really the real deal in terms of drug discovery. What we hope is that therefore this will accelerate the translational aspect of the work that we are doing.