Past Year as a Junior Bioinformatician in Perspective

Starting a new chapter in life this week attending grad school (again) towards a PhD. Before moving on, wanted to put into perspective and share a few notes on what I’ve learned over the past year working as a junior bioinformatician for a service core. It’s been a long journey, was a difficult decision to leave, and have moved several different institutions and cities now. I am focusing and following my own research interests. In the long run, I’ll be of better use to people this way. Looking forward to the next steps!

  • I got the job by responding to a post on the Bioconductor listserv. I had an interdisciplinary-enough background coming with MS research degree, 4 years of molecular bench work experience, some makeshift computational/programming skills, attended excellent Workshop on Genomics, Software Carpentry workshop, and was really interested. They were willing to train me.
  • They had foresight to hire junior-level positions. My colleagues and I did the more routine stuff like Illumina demultiplexing and fairly well-worked out pipelines, e.g. RNAseq so the higher-level bioinformaticians could be free to tackle larger research projects and administrative challenges.
  • Worked on a side project benchmarking de novo transcriptome assemblers, ended up being in the right place at the right time, and wrote a review chapter for a book.
  • Learned to reach out to people I admire and ask questions.
  • The perfect is the enemy of the good.” (or as someone I met recently put it, why does it have to be good, can’t it be excellent???) It’s never going to be perfect. Do your best to get results out, otherwise they will sit with you on and on for continued time.
  • I worked on data analyses for 17 different PI in the past 12 months. Some in more depth than others, but ultimately had to learn about 17 different research goals/questions/designs. That was just me. My colleagues did even more than I did.
  • There’s no substitute for working in the same office as two nice, smart people working on similar things. Or walking across the hall and asking my supervisor, having him spend the time to outline an answer or discussion so that we both understand more about what’s going on. I’ve motivated a lot of self-guided study over the past few years, learning to seek help through the online community when I was isolated in a place where no one else was doing what I was doing. But, if there’s a problem that I’ve been staring at for the past 20 min-to-an-hr, having people physically present and just asking “Can you take a look at this?” can save a lot of time and has really escalated my understanding and grasp of bioinformatics. Plus, it’s fun to laugh together when service clients send funny emails. I am eternally grateful for my former officemates, supervisors and all of my colleagues.
  • There was genuine interest and initiative for improving relationship between biomedical research community and bioinformatics support. High demands for diverse services create challenges and constraints on time.
  • There are benefits to being at a large institution (I had come from a small academic institution). If someone doesn’t know the answer to your question, someone else might. Or, different conversations with different people will be thought-provoking towards a different way of looking at things. Very supportive, diverse and productive atmosphere.
  • Co-author on 2 papers, RNAseq and 16S community profiling analyses
  • That’s not to say there aren’t politics associated. There will be negatives about every place.
  • At the end of the day, you have to be happy with both what you’re doing and where you live.
  • In a service-oriented facility, some PI just want results. I’ve experienced this stalemate first hand:

  • Bioinformatics is a process, way of life, a field of study in itself, and not just a set of software tools that will output results. Requires hypotheses and protocols and being skeptical of your results then going back and doing it again to make sure you can reproduce. What I love about bioinformatics is that you can go back and do it again in a slightly different way to see what happens, testing what parameter caused the interesting results. Not like in the lab where reagents are SO expensive, samples hard to come by, that if you were to go back and do it again, it would double your costs. With bioinformatics, there are costs but in time and computing resources.
  • I’m convinced that results and interpretations of bioinformatics tools are better if the PI or grad student or someone deeply involved in the research question learns about the tools, can program in R or use the commandline tools themselves. I really think that every biologist should learn how to program, but I realize that’s not realistic expectation. So, at the very least, if bioinformatics results are a necessary part of the study, students, post-docs, someone in the team, should be active in learning about how to use the bioinformatics tools.
  • I knew stuff about microarrays and was willing to analyze microarray data. People still have microarray data and seemed to be grateful. The field seems to be moving away from microarrays. Yes, data are messy, noisy. Don’t give you unsolicited sequence information like RNAseq. Cost of microarray vs. RNAseq might even be higher at this point because of analyses compared to RNAseq. But, I’m not convinced that this is a completely dead technology that should be discarded. Might be more appropriate data type for some questions. Such as…?
  • I’m glad that I postponed grad school over this year rather than accepting admissions offers last year. It gave me a chance to explore more in depth, and meet new people. Question the grad school process and even apply to a few more schools. (Another post forthcoming about grad school application process.) Now is definitely the time. Answer to ‘Do you really want a PhD?‘ is now a resounding “Yes!” Can’t imagine doing anything else. Excited about future.
  • I still feel like a gigantic moron most of the time, write code badly in a way that is not collaborative or sustainable. Sometimes I find that I can’t reproduce my own analyses several months later and then end up having to re-do past work. (Everything is documented, just in a haphazard way that makes sense for me, but not necessarily someone else – or my future self in 3 months.) Need to work on all of that that. Session on reproducibility at NGS2015 helped tremendously as does github.
  • I still remember the first time I got a big HPC job running, and saw my name on the board. It was exciting! Don’t forget that:


About Lisa Johnson

PhD candidate at UC Davis in Molecular, Cellular, and Integrative Physiology
This entry was posted in Bioinformatics, Data Analyses, Personal, Sequencing. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s