In medical research, what is the relative value of confirming old work vs. new discoveries?

Question

I am an MS student in medical immunology. I am looking for my research proposal, and would like to get a nice publication (i.e., publish in a journal with a high impact factor). Given this, is it better to do something totally new, or is there value in repeating old work?

I ask because my supervisor is busy and gives me considerable freedom. I have the idea of experimenting with the role of some gene in an autoimmune disease but no one has done that work before. Is it good to pursue this? When I write it up, how will I do citations?

Answer 1

Note, I have a physical/computer sciences background; things may be different in life sciences.

I have the idea of experimenting with the role of some gene in an autoimmune disease but no one has done that work before. Is it good to pursue this?

Yes! Doing novel things will get more attention than revisiting old things (though both have value).

My concern here would be whether you, as a lone, beginning student, have the background and resources to do something completely novel. Normally, one's first study is done under the close supervision of a mentor. I'm a little surprised your supervisor is "turning you loose" like this.

When I write it up, how will I do citations?

Presumably parts of the methods you used have been used before (perhaps for other diseases), and someone has studied this disease before. So, you should cite any past work, even if it's not a direct hit to what you're doing.

But, this is sort of what I meant above -- it's good to have a mentor to help walk you through this process for the first time (or first several times) -- getting novel, publishable results with no supervision and little experience seems like a very tall order.

Answer 2

From a research standpoint, if you repeat old work and get the same result, then you haven't advanced the state of knowledge and so probably won't be able to publish anything more than a note.

One exception would be confirming an older result but with a completely different methodology. Then it is the new methodology that is the "new and exciting" part that might merit publication.

But, especially in the medical field, if you repeat old work and show that the original work was wrong, or even dangerous, then you have something worth publishing and can become famous (or infamous if you are wrong). But I think that as a student, you should undertake such work only if you have a strong suspicion and some evidence that the original is in error. Otherwise you won't gain much from just following other's footsteps.

So, in general, it is probably better to work on untrod ground, on something new, as hard as that can be. The fact that your advisor gives you a lot of freedom is good, but only provided that they also give you guidance on your ideas and also provide some worthwhile ideas for you to follow.

I'm not sure what you mean by your question on citations, but it is typical to do a "literature survey" before you start on a new research direction to make sure you know what is known about the topic. If things are already known (likely), you will cite them if you build on them in any way.