Dr Hamish McAllister-Williams: The three Ps, person, place, project are all critical, but I think more people fall down on project than anything else. And I think the major reasons for that usually fall under the headings of either feasibility or scientific rigour. Feasibility though has to be the number one because if it’s not feasible there is no point in the project. Sometimes with regard to scientific rigour, if there is good training and it is more blue sky, it may still be okay. But something which is scientifically completely unsound will not get funded. Professor Eileen Joyce: Key things that I look for in a successful application are clarity of exposition of a subject matter. Because we take applications from a whole wide range of medicine, and the panel members tend to be in specialist areas, they often would be reading applications that they’re not a specialist in. So we need some sort of clarity to understand the content of the project that they’re proposing. So it’s clarity, succinctness, simplicity of experiments, not too many experiments, not too many aims and objectives, and the experiments should be doable by a single person. Sometimes we see people who are really proposing studies that are almost akin to clinical trials which would need the back up of many other researchers, as in clinical trials, and they clearly are not doable in a research fellowship like this. Professor Moira Whyte: So in terms of the project, we’re looking for a clear hypothesis or research question, a good rationale for why that question needs answering, ideally in a relevant clinical context. And we’re looking for a well-structured series of experiments that may be inter-related, but the results of the whole project are not entirely dependent on one experiment. We’re very keen to see a plan B, particularly with an innovative experimental approach, what if it doesn’t work out as you will expect, do you have a fall back plan? Professor Neil Hanley: You have to look and try to imagine if this project goes well, and everything works well, what will happen? What will come out of it? Because if the very best that would come out of it is a middle ranking publication, that isn’t going to launch your career. Whereas what you’re looking for is if this project goes well it really could launch your career with big impactful outputs. Professor Jane Armitage: Common issues that we see are problems with having not thought through things like sample size. I think we see it both in the animal experiments where people pluck numbers, “We’ll do this in six mice or ten mice,” or whatever, without thinking carefully about how many mice they might need, you know, to detect particular things. Similarly, in clinical or human experiments, people not thinking carefully about the statistics behind what they’re trying to do. Sometimes they will have gone to a statistician asked a, sort of, simple question, got a simple answer, but actually the question was not framed in such a way that they got the answer that they should have got. So I think there are a lot of difficulties there, and I think people do need to think about the numbers behind what they’re doing. Professor Ian Sabore: The thing that I think that I see that is most frustrating is people who don’t use all the space to really demonstrate and think through their project. They might come up with a reasonable idea, and then there’s two or three methods that might be applicable. But they don’t necessarily always then really explain the experiment or show that they’ve grasped the methodology and what that methodology can deliver. So I think I would say use the available space, keep the references short and snappy, take the methodologies that you want to use and show why they’re the right experiments, and that you’ve thought through it in detail. And that you can then defend that experiment at the interview and say that, “This is the experiment because… It’s
deliverable because… I know that I can do this methodology, it’s been done in my lab, I know we have the right cells and the right resources because here they are and I’ve thought about this issue. There might be a problem here, I’ve thought about how I’m going to handle it.” Professor Paresh Vyas: The most common issues that the case for support isn’t well argued, there isn’t a clear question, there may not be a clear set of observations, and the methodology may not be appropriate. Time must be spent on the case for support. Dr Hamish McAllister-Williams: I think the big challenge for the case for support is the amount of space you have to be able to write your case for support. And it is also true that many referees say there is insufficient information in the case for support, and yet it is possible to put in a very good case for support in the space available. But to do that it means not wasting too much space in background, you have to able to contextualise the research, and the scientific background support does have to be nailed down, but you need to do that, I think, fairly briefly. Because, I think, the majority of the space needs to be set over to the description of the actual project, what is being done and how it is being done. That is where people will often slip up because there’s just insufficient information for a referee or the panel to be confident that
the study is both feasible and scientifically sound. And I think there are one or two very specific elements that it’s very important to make sure that are done correctly, and that would include things like power calculations, there has to be a clear justification for the numbers of subjects included, be they animals or be they humans. And that again is something that is sometimes not always done as well as it could be. Professor Eileen Joyce: Well, there’s not very much space for the case for support so it’s critical to get across the experiments that you’re proposing
to undertake, and to justify them. Sometimes we find that candidates spend too much time on the background and too little time on the details of the experiments that they’re proposing to do. And that often lends to lack of confidence that the candidate really understands the kinds of experiments that they’re doing and have worked through them properly. Professor David Ray: So in terms of the design of the case for support, it’s not sufficient just to outline in general terms some approaches that could be taken. I think it’s very important that the fellows consider the variability in the measurements that they’re going to be undertaking, what potential confounders and bias there may present in their experimental design. Professor Ian Sabroe: Everybody, I think, worries about preliminary data, but the thing is the panel actually also looks at where you’ve come from as an individual. If you’ve had time to make preliminary data and can show that you’re thoroughly adept in a key technique, then obviously that’s lovely, but not everybody will have had those
opportunities. So what you have to do is to show that you have engaged in your project to the best of your ability for the stage of the training that you’re at and with the opportunities you’ve had. If you can put in some preliminary data that you’ve generated, that’s wonderful, if your lab has some that is unpublished that you can use and attribute straight forwardly, then that’s also very useful. But you are where you are, and the panel understands where you’ve come from.