I'm an evolutionary and computational biologist doing my second postdoc at Michigan State University. I have learned all sorts of thing in this short career, and Jeremy Yoder has asked for advice for a new blog canival.
If I knew then what I know now, I would have...
Focused more on my writing skills very early on. At least as early as my graduate degrees at KGI and UCSB, but perhaps I should really have gotten more into writing during my undergrad in Copenhagen. No one ever told me that being a scientists really amounts to being a writer. I have done nearly nothing but reading and writing for at least six months now, save for giving some talks at meetings and writing 32 lines of code. Write even if you have no data and no conclusions. Write your thoughts down on what you read, what you do in lab, and then it will be easier to write the thesis and papers when it really counts.
Read more. As a scientist, reading is treading water. If you stop, you drown. It's a never ending game, and it is the only way to keep abreast with what is going on. Going to talks is fine, but simply not enough. You must read constantly, or you will be left behind. Often it just means reading abstracts, sometimes also looking over figures (and reading captions) - not that I count, but I count reading abstract an figures as having read a paper. Sign up for eToCs from the major journals in your field. I recommend: Nature, Science, PNAS, Proc. R. Soc. B, Genetics, Evolution, Journal of Evolutionary Biology, Frontiers in Evolutionary and Population Genetics, PLoS Biology, PLoS Comp Bio. After my first year in grad school, I read the advice from a senior scientist that one should spend the entire first year of grad school mostly reading. I did read a lot, but wish I had read more.
Stopped taking myself so seriously. Actually, I haven't done that in years, but I do think this is invaluable advice. It's just science, after all. If I am wrong about the prevalence of epistasis in adaptation, nobody is going to care. No bridge will collapse and no one is going to die of a misdiagnosis. Keep that in mind, and enjoy yourself. Unless you're an engineer or an M.D, in which case you should stop reading this blog and get back to fukcing work already, or I'll sue your ass off!