Professor David Gavaghan – Professor of Computational Biology, University of Oxford [DG]
Mathematics is increasingly needed in biology because what we are trying to do to address some of the most pressing questions in medical research in the 21st century, turn biology into the same sort of predictive science that physics is already. In general what we are trying to do is understand how the heart works in a predictive and quantitative way so that we can actually write down mathematics that describes how the heart works and then use mathematics to predict how the heart is going to behave in the future. In particular, what we would like to look at is whether a new drug that’s potentially being introduced by a pharmaceutical company, is likely to cause side effects to the heart which are dangerous. I think everyone has heard of various drugs that have been brought to market that eventually caused cardio toxic side effects and even killed people.
Professor Denis Noble – Emeritus Professor of Cardiovascular Physiology, University of Oxford [DN]
The heart has two major aspects to its mechanism. Most people will think of it as a pump that moves blood through the body and that is its primary function. But how is that controlled? It is controlled by electrical changes in the heart itself. That process of the electrical waves spreading over the whole heart is actually generated by many tiny objects that we call miosites, that is muscle cells, and each muscle cell has a membrane around it and in that membrane there are proteins that carry charged particles which we call ions – sodium, potassium, chloride and other ions – across from one side of the membrane to the other. That is what carries an electric current and it is the carrying of that electric current that generates in each cell a wave and then that cell communicates with the next cell that communicates to the next one and so the wave passes over the whole heart. Every time it passes over each cell that electrical wave triggers the cell itself to contract to produce a small movement and when you add up all of those small movements you have got a whole heart beating away.
What we are interested in is how a drug will interact with an individual muscle cell and change what we call the electrophysiology, the way the electro citation spreads through the heart, and what we can do is mimic what a cell does in mathematics. There is software that runs to keep aeroplanes up in the sky that make sure they don’t crash into another and that software is written to very high specifications, it is called safety critical software and its safety critical because if something goes wrong, then a catastrophe happens. Medical and biological software that’s modelling the way processes happen in the body and might be used for drug design, ought to be developed to the same high specification if we are going to use it to make decisions about individual patients. It’s got to be right and we have got to be absolutely confident in the software.
Dr Gary Mirams – Senior Research Fellow, University of Oxford
We use computational models to put together information from different experiments that the pharmaceutical companies do. They can screen how much a drug compound blocks particular individual channels and then we use the computational model to put that information together to make a prediction for what will happen to a whole cell or a whole organ.
The development of a new drug, whatever it is for, is a very long and costly process. If a drug fails at phase three in clinical trial, so the very last stage, the drug company would have spent something between 500 million and a billion dollars on getting that drug to that stage and so that money is in effect wasted if that drug doesn’t come to market and that is one of the reasons that drugs cost so much. People like GSK and AstraZeneca, we also work with the food and drugs administration in the US, are happy to trust our code because that’s the way it’s been developed. It is also completely open source so everybody can look at it, you can come onto our website and download it yourself and there are lots of tutorials.
In terms of taking the multiple actions of drugs seriously and computing them, the answer certainly is the virtual new age because up till now what is tended to be the way of looking at a problem with drugs, is to say they have got major action, that’s the main target and all the other effects are side effects. We are saying, hey no wait a minute the other effects may be very important and it is in combination that you can understand the action of your compound.