I know this sounds like a bizarre X-Men comic book issue, but hear me out.
Being a wet-lab researcher in biology or medicine for the past 20 years couldn’t have been more exciting. Massive amounts of biological datasets have been generated to better understand the health and disease state of the individuals. As a researcher, this hopefully means that hundreds of hours of work in the laboratory will lead to exciting achievements. However, your precious experiment result is waiting to be analyzed. Who is going to do it? The bioinformatician? Some people would like to think otherwise. I would like to think otherwise.
As a freshly-started PhD student in Dr. Nathan A. Lack’s laboratory at Koç University, Turkey, I was eager to take on this challenge: I was going to be a wet and dry-lab rat. The advantages were enormous. Freedom of analyzing the data with the non-existing “gap” between the biologist and a bioinformatician and, therefore, avoiding endless meetings. Even adjusting the color palette to my taste without kindly requesting looked – liberating. These all sounded amazing; however, I did not know where to start. I had always been interested in computers and considered myself a “power user” with a fundamental understanding of web applications and object-oriented programming. Still, my bioinformatics toolkit merely consisted of basic commands in the terminal. I needed a boot camp – a head start.
To improve these fundamentals, gain more insights about working with cancer biology datasets, and start doing in-house computational analyses, I attended a Cancer Genomics workshop at the European Bioinformatics Institute in Hinxton, United Kingdom. Most participants were solely computational, and a minority were experimental researchers. I remember clearly that I was trying to explain the benefits of being self-sufficient to a wet-lab scientist to convince her to analyze her own data. It wasn’t easy. However, we agree that we will do whatever we can to make this happen. All things considered, that 3-day training clearly put the puzzle pieces of certain bioinformatic analyses into a more realistic context. Starting with publicly available datasets, I followed tutorials, talked with fellow bioinformaticians, and asked questions to authors of the bioinformatic tools at every chance I got. I spent hours understanding the logical connection between assumptions that algorithms make in biological contexts. In the meantime, I probably ran meaningless analyses in university’s high-performance cluster computers. After spending a great deal of time working on these analyses while I conduct my experiments, at the end of my PhD, the result was encouraging: I could reproducibly analyze different types of biological datasets – including my own.
Other than the technical side of things, I felt a particular social perspective about working on “both sides” of the research. My experience in both domestic and international research communities was almost the same. According to most people, when you finish your experiment, you’ve done your job. A better-suited person in your laboratory or department will probably analyze the data more quickly and accurately than you. Only then you could start working on the results. That is perhaps true; an individual who performs these analyses regularly will be way better and faster than you are; however, one has to start somewhere. Furthermore, If you are primarily a wet-lab scientist, you will never become “computationally savvy” as the people in bioinformatics will never think of you as one of them. On the other hand, for wet-lab people, you’re just a wannabe bioinformatician that plays around with the data and will eventually rely on other people to finish the analysis. Although it is unreasonable to specialize in every subject, the science is multidimensional, and I think researchers should work as much as possible to deeply understand every bit of this vast picture.
Over the years of my PhD training, I gained some insights that might help other fellow researchers that want to be a successful, self-sufficient scientist that can generate and analyze their own data.
Be eager and brave
You’ve listened to all sorts of computational tools, analyses, and results, but this is a new realm for you. Be prepared for the frequent head-scratching and asking fellow bioinformaticians. Offer to participate in “computational-people only” meetings.
Don’t reinvent the wheel
There is already a multitude of information out there: communities such as Software Carpentry and courses like Missing Semester or books like Bioinformatics Data Skills will build up the foundations of your computational arsenal. Stack Overflow and Biostars will be your best friend.
Engage with fellow researchers
It is odd, but I found it extremely helpful to get into conversations with wet-lab researchers about any type of computational analysis. You will come up with more concrete questions at the end of these discussions. Doing this also will boost both your and your colleague’s motivation. Building communities and meeting for casual “workshops” or troubleshooting with your own datasets will sharpen your skill set.
We’re heading to a one-way path where wet-lab researchers eventually form this new-generation hybrid scientists. It is actually happening as of now. I met with tens of researchers ranging from medical students, graduate students and early-career scientists that they are not just interested in doing this but they are already doing it. I am also aware of the convenience of the position of the wet-lab scientists compared to the dry-lab scientists such as bioinformaticians and computational biologists. Being mainly a dry-lab researcher who wants to start doing wet-lab research is much more complicated than vice versa, although it is not impossible. All-and-all, this is my humble recall to “not leave the biology out of bioinformatics” (Campbell, 1999) as “biologists are the bioinformaticians now” (Stein, 2008).