Monday, October 6, 2014

Unsolicited advice for new graduate students in the biological sciences

School just started up again here at UC Davis, which always makes me reflective about my first days as a graduate student. I've had a half-formed post in my mind for a while now about things I wished I'd known when I first started, so I thought this would be a good time to get them out there.

My experience is probably most relevant to people in the biological sciences, and especially those in microbiology. The further you are from that, the lower the chances that this might apply to you. But if your're reading a blog titled "Mostly Microbiology," I figure this might be up your alley. So without further ado, here is my list of unsolicited advice for new graduate students

Bad dad joke


1) Finding a lab/advisor.
Your number one priority in your first year. There are several things I think you should care about in finding a lab:

A) Personal Fit. Finding a fit between a advisor/professor and grad student is a funny thing. Different people care about different things. First, ask yourself how self-motivated you are. Be honest, as this helps to guide your selection. Do you want a mentor who will just leave you alone to do your work without interference, or one who leans more toward active mentoring? There isn't a universal rule as to which one is better, and many professors will be more active at the start of your time with them and wean you off their help over time until eventually you are working more independently. Just know that if you join a large lab with a very busy professor, she or he may not have as much time to be involved in your projects. Professors are busy people. You may get left in the hands of senior lab members (postdocs or staff) to answer many of your questions. Often times these are very capable individuals, especially with the current glut of people stuck waiting around for their own tenure track position to open up. Twenty years ago people with their type of experience and talent would have likely had their own labs already. The positive trade off with having less face time with the professor is that larger labs are often a sign of funding success (see point C).

Success baby is living the dream

B) Research fit. Ideally, it is better to find a lab doing things you are interested in. This will likely be your life for ~5+ years. Be open minded with rotations, as many subjects might not be immediately interesting at first glance, but reveal more intriguing aspects upon further study. Do you want to do basic research about the fundamentals of how things work, or more applied research that directly leads to a product or a cure? Do you want to do medical-oriented research or environmental/ecological research? The more options you have to choose from the better chance you have to find one that has funding for you. Speaking of which...


C) Funding.
At least not on the timeline one wants


Few things are worse than running out of funding halfway through your degree. Many principle investigators (PIs, so called due to their status as the person listed as getting the grant money) don't know if they have funding until right when you need to decide on a lab. Lab funding is a touchy subject, and often PIs have to adapt on the fly to changing circumstances. Funding yourself by landing a teaching assitantship (pays your stipend and tuition so it doesn't come out of their grant money) helps. It's a balancing act however, because the more you teach, the less time you have for research. As an aside, I think it's a very bad idea to have any form of outside employment while in grad school.


2) Classes


NOBODY

This is more controversial. In my opinion, your goals with graduate level courses should be to a) learn all the stuff relevant to your goals and b) pass the class. Grades are secondary. I've not seen very many places that care about your graduate GPA if you have a degree and publications. I was even told that it was tacky to include my GPA on my CV. Some graduate fellowships ask you for it, but from what I've heard, as long as you aren't failing, they mostly care more about your research proposal and not about your GPA. It's not like a 3.0 vs a 3.5 is going to make or break an otherwise outstanding fellowship application. Don't sacrifice research time and sleep to break your back for that A+, when you can achieve a B for half the effort. This is a very different approach from your undergraduate days, and lots of high-achieving people struggle with it. But the cost/benefit ratio (in my opinion) lies firmly on the side of more research and a lower GPA.

3) Research


True story bro (if somewhat hyperbolic)

Don't expect research success right off. Nearly everyone screws up a lot at the beginning. Here's a few general principles that might help you to not waste too much time and effort:

  • Plan out your whole experiment, timeline and all, before you get out a single tube or reagent. Make sure you have all the necessary supplies. Few things are worse than realizing halfway through that you don't have time to finish that day, or are out of some critical reagent, and having a bunch of effort go for naught.
  • Think about what controls to include. Doing the experiment the right way the first time sure helps.
  • Try to design experiments that will tell you something useful no matter how the results turn out. Hoping for one specific result will inevitably doom you to being cursed by the Hacker Gods who programmed this reality (and like to mess with us from time to time) to uninterpretable results. Ask yourself what each possible result of an experiment would mean, and if the answer is "I don't know" or "Nothing," tweak your study design if possible.
  • Design a physical workflow that helps you keep track of what you have and haven't done to each sample/tube. I like to move each tube from one rack to another (or from one row to another) after adding a reagent so that I know that I have added it to that tube. That way, if I get interrupted, I can go back an know where I was at with each tube.
  • Learn some basic programming skills. I am still working on this, but it has been incredibly useful, and is a huge time saver in the long run. I like python.


The general idea with your research is that you should eventually know more about your narrow project than your mentor does. She or he will of course have a much larger total amount of and a wider breadth of knowledge, but you should know the particulars of what you are doing in and out, up and down, inside and out. Read ALL the relevant literature.

Ok, so that's pretty much impossible. Unless you define relevant tightly.

Figure out where you think the field you are in has things wrong, or where there are important holes in understanding. Once you have proven yourself with some initial success, start trying to branch out from the initial path your advisor put you on (however specific or general that path was). Start making suggestions of your own about how to approach the important questions. Isn't that kind of the point of grad school, to develop the ability to do independent research? You want to become a colleague of your advisor's, not just an employee. Of course, don't spend supply money or too much time on things that you haven't gotten approved by your advisor. They are paying you, after all. Just be an active participant in your research.

This is a process that takes time, of course. Meanwhile, read about, and attempt to banish any impostor syndrome that might set in. I like these two articles to understand the phenomena. It happens to lots of us, myself included.

I think just about all grad students feel this way sometimes.

Lastly, it is never too early to think about what you want to do next. Your advisor is probably most familiar with the academic route, since it is the one they took, but there are lots of other careers out there. Science communication, business, law, non-profits, government agencies, and public policy all have niches where you might find a home. I know lots of people who started grad school because they had a "get ALL the degrees" approach to life, and that's fine, but there are things you can do now to dip a toe in different ponds and see how you like it. Take advantage of opportunities to try on different shoes when they come your way in department emails, workshops, or volunteer opportunities.

Good luck this year!

Monday, September 8, 2014

New Paper: Stool Microbiota and Vaccine Responses of Infants


As part of getting my blogging groove back on, I thought I would talk about my part in this new paper. Together, Nazmul (the first author) and I analyzed the gut microbiota of a large cohort of infants in Bangladesh, paying extra attention to the bifidobacteria. Nazmul (and others) did a ton of work looking at the response to various vaccines in these infants. We found that the response to some vaccines was better in infants that had higher levels of bifidobacteria.

A couple of things of note about my contribution to this paper:

1) The correlations between vaccine response and the microbiota sometimes differed between species and subspecies of bifidobacteria, which means that A) not all bifidobacteria are equal, and B) that the effects can be correlated to a manageable set of genes. We were able to trace the effects down to subspecies level within the B. longum group thanks to a new method that I helped develop, which should be getting it's own paper soon.

2) The levels of bifidobacteria in these infants were really high compared to some other cohorts I have seen or studied myself. This was despite the high rate of c-section birth noted in the paper. I find this super interesting.

3) Parents, please vaccinate your kids so that stuff like this and this doesn't happen. Herd immunity is a real thing. When you choose not to vaccinate your child, you put other kids who can't get vaccinated for whatever reason (immunocompromised, etc.) or for whom vaccines don't work at risk. No matter what type of bacteria live in your child's gut, vaccination is almost assuredly better than no vaccination.

Anyway, I was glad to use my Bif-TRFLP technique on a new set of infants and be involved in this cool study. Grad school has given me the chance to collaborate on projects with a lot of great people on interesting things. I'm glad Charles Stephensen got interested in gut microbes and started this collaboration with my advisor. I'm increasingly convinced that collaborations are the best way to do impactful research. It's so hard to be an expert in all the relevent aspects of today's unsolved problems.

New blogging style

Life has become very busy lately and I have sadly neglected my once-thriving blog. I've had a lot of writing to do lately, and haven't felt much like adding more to my plate. To remedy this situation, I have decided that other than the occasional long-form post about new papers, I am going to try to write shorter quick-hitting posts from now on. We'll see if that helps lower the activation energy and gets me posting more. I do enjoy expressing myself in my own forum, but writing long posts became too daunting.