I'm a first year post doc in bioinformatics and am currently working on an application project based on other's method previously developed in the lab. The application would simply be applying the existing algorithm to a biological problem and validate with experiment data. Since I'm interested in a academia faculty position, would this diminish my chance (potentially due to novelty issues by simply cashing out others work), or should I suggest to my advisor that I would rather work on other project based on my own ideas and develop my own methods? Does application of an existing method counted as novel by the academia? This project is assigned by my current PI and I'm constantly worry that I will be viewed as unoriginal by working on others method. I had to ask because my PI keep me piling up more of these "case studies" recently. I just feel that my PI care more about successful applications of this existing algorithm. If I did a great job on these projects, would this be helpful to me academically? I just want to make sure before I accept more of these cases or maybe talk to my PI so I can work on a novel direction. Thanks for advices/suggestions.

  • 1
    Can you talk about what field you are in and give us some more details? The answer is going to vary wildly. – Fomite Jan 16 '17 at 8:29

As a general rule: publications tend to be the measure of whether something is novel. So, if you are publishing, I'd say you indeed are helping your academic career. If you aren't publishing: that is a major complaint you can bring up with your PI. Here you can use the outsize importance of publications in academia to your advantage. You can debate all day long whether your work is novel enough with your PI, but no one can justify leaving you without publications.

Also, it is important whether these applications are helping you grow as a researcher. Are these applications trivial, mindless work, or do you learn things while working on them? Are you getting faster/better at implementing them? Are you ever surprised at results, or can you predict what the results are going to be before the experiments are done?
Note that breakthroughs in science often rely on a strong grasp of the fundamentals and the context in which the work is done. Having strong experience in applying those algorithms could be essential in later designing them yourself.

  • Very true but there is only one waterman-smith but many BLAST users! I think eventually I will need to work on something novel. – user67765 Jan 16 '17 at 9:51
  • I can't judge whether you are given the challenges and fundamental problems you could reasonably expect. One way to get involved with more fundamental research regardless is to look at the algorithms you are applying and think about what you miss in those algorithms. What problems do you have when applying them? What properties are properties that the algorithm doesn't have, but would make a big difference in applications. Then approach your PI or the people that are doing more fundamental research with these ideas and ask them about those issues. – dimpol Jan 16 '17 at 10:05
  • Thanks, I got some ideas and already did. Its just that I got too many of these case studies that my PI eagerly want me to prove this existing algorithm works. I really want to get these out of the way and get to the sweet stuff. – user67765 Jan 16 '17 at 10:14
  • 1
    Then I'd advise you to have a 'bigger picture' conversation with your PI. First talk about how you want to do more fundamental research. Maybe even talk about how the applications made you enthusiastic about working on the algorithm on a more fundamental level. Then discuss the length of your post-doc and what percentage would be the right amount to spend on applications and what percentage on more fundamental research. Then ask about how you can get involved in more fundamental research AFTER your current responsibilities in applications. – dimpol Jan 16 '17 at 10:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.