Meet the panel
Meet the panel
Meet the team answering your questions this week: Sarah Harrison, Simon White, Peter Clarke and Olivia Remes.
Chris - First, let’s meet the panel: Sarah Harrison, Biologist, University of Cambridge - Sarah, you work with stem cells?
Sarah - My specialism is about using stem cells to try and make organ like structures and embryo like structures in culture, so instead of being in the environment of an embryo itself. Stem cells have quite a lot of capacity to do this pretty spontaneously. If you leave them to do their own thing they’ll make neurons by themselves, but what they don’t do very well is make these structure in a reproducible way.
Chris - But why do we want to do that at all? Why do you want to grow mini bits of body in a dish?
Sarah - It would be useful to study how they develop by using these models rather than using real embryos, but it’s also quite a good platform to start to test drugs as safely as possible.
Chris - So you could give your drugs to a petri dish rather than a person to work out whether or not a drug you're experimenting on might or might not work?
Sarah - Certainly with some organoid systems drug screening by using chemical compounds and putting them in dishes would work, yeah.
Chris - Thank you Sarah. So any questions relevant to biology, stem cells, and embryos should go Sarah’s way.
Simon White, Statistician, Biostatistics Unit also at the University of Cambridge. Stats, that’s a word that makes people shudder and often run a mile Simon, but it hasn’t in your case, you're a professional statistician. So have you any examples of uses and abuses of statistics to impart with?
Simon - I have. I think statistics is one of those words that instills fear, but really we should just think of it as the idea of trying to understand what lots of numbers can tell us. I think statistics shouldn’t really be about formuli and complex numbers, and lots of random numbers like oh, the number of people in this room is 5. But what does that tell us about the people in the room - nothing much. That’s what statistics is about, understanding what a number tells you, rather than the number itself.
Chris - It’s pretty important though,isn’t it. Because when we can start to ask questions about very large numbers of people we can learn, for example, what things cause certain diseases, or what things protect you against certain diseases, but you’ve got to look at lots of people to make sure you reach the right conclusion?
Simon - I might correct the word “cause” in what you just said. Sometimes a lot of them are associations rather than causal studies.
Chris - You can tell we’ve got a statistician in the studio. So anything to do with maths, and stats, and that kind of thing should go Simon’s way.
Sitting next to Simon is Peter Clarke, who is the founder of company RESURGO genetics. Peter, what is that?
Peter - What we’re trying to do is use some of the latest advances in machine learning that maybe we’ll talk about later, and really try and understand how cells work and how cells talk to each other and, through their communication, make you.
Chris - Machine learning, what do you mean my that?
Peter - This is really taking computers and getting them to look at lots and lots of data and trying to find the similarities, so seeing the same structures over and over again in data. Once you’ve got that, you can start building up pictures of how things really work even when you maybe don’t know ahead of time what they do.
Chris - So the computer essentially teaches itself to see the relationships between if it looks at lots of things that happen, it can begin to see what happens when this happens and it can begin to, therefore, may inferences, but you don’t know how it’s doing that, it’s just learning by looking at lots and lots examples?
Peter - In the same way that we don’t really yet properly know how your brain works, for example. You’ve managed to survive in the world though lots of examples and trying to imagine the the ways of doing things. But, just because it’s a black box doesn’t necessarily mean it doesn’t work.
Chris - What would your company sell then? What would I come to you asking to buy of RESURGO genetics?
Peter - That’s a good question and one that our investors are asking us a lot.
Chris - Is there a business model?
Peter - I think the business model in the long run is if you develop a far more sophisticated view of how these cells actually work, and how they communicate to make you that this becomes a very useful tool in medicine, and agriculture, and all sorts of other things.
Chris - Sarah’s trying to grow cells in a dish, you’re actually growing cells in a computer’s memory, effectively?
Peter - There’s a lot of overlap between what she does. We’re almost the computational modeling side of what she’s doing.