gravatar for ATpoint

1 hour ago by


You should get a supervisor to give you a project that is managable in the time you have available while neither over- nor underdemaning you in terms of your current skillset. You probably have a few weeks time, try and see whether you can formulate a project to get the basics of analysis right. I do not want to discourage you but do not expect that as an undergraduate you are going to make any notable discoveries which are new and interesting, for this you simply do not have the time, plus if datasets are available for download then most likely someone already looked at them and based a research project on them. Undergrad projects should be tailored to learn something, improve yourself, not making discoveries, that is for PhD or Postdoc projects where people have the time, skillset (which they may acquire over time) and a lab in the back to generate and validate datasets as needed.

RNA-seq itself is commonly used for differential gene expression analysis, that is which gene is more highly-expressed in one versus the other condition. Or to find new isoforms of a gene (humans are outstandingly-well annotated, unlikely to find true novelties plus requires experimental validation). Or differential isoform analysis (which transcript per gene is more highly-expressed), more difficult to interpret than differential gene-level analysis, plus again requires validation and additional data to correlate with, e.g. clinical outcomes etc to see whether this has any implication. Or variant calling so finding mutations in exons, not the ideal technique for it but possible, again requiring validation with independent experiments.

There is a lot that can be done, but the research question should come first, and one then chooses experiments and techniques accordingly to answer it. Doing it the other way around is cumbersome if you ask me as you force a certain dataset or technique into "giving you a research question".


modified 1 hour ago

1 hour ago


Source link