Those were the charming days, as we often view our past and curse the present. I was working in a biochemistry lab performing enzyme assays tirelessly. I often had to get up in mid night or early morning hours and run to the lab to pursue the time-course enzyme assays. During the breaks (incubation/treatment of samples), I used to follow a senior colleague of mine to learn RT-PCR, gel runs and my favorite part the 'primer designing'. Primer designing led me to further explore the area of sequence analysis and bioinformatics in general. I took a short online tutorial offered by Manchester university. I find it appealing, but my fear was 'coding', so I never viewed myself pursuing bioinformatics as career option. It was a mere coincidence of an honest advice and a sudden movement of my then PI to another city, which allowed me to re-set my career trajectory. The advice was straight forward. "many are doing RT-PCRs and enzyme assays, how many biologists are coding? The ones with knowledge of biology as well as coding can do something that none of others will be able to do. If you like bioinformatics, go ahead and learn programming. You will have the greater market value". I took this advice seriously, followed my interest in bioinformatics, and worked day and night to learn coding. Eventually, I became a bioinformatician, however with a cost attached. The cost was the label of "theoretician".
Once you know the coding, people will grab your neck to solve their problems and twist your arms to teach them what you know. Asking them to teach you what they know is a criminal offence because: i) They knew that their skills were pervasive and yours was unique and little scarce, ii) they consider you a programmer, not really a scientist leading the projects, and they like this arrangement. iii) They presume that being a computer nerd you will not be able to hold eppendorfs and pipettes, and it is going to be time -massacre for them to teach you. iv) They will be jealous of you sipping coffee at your computer desk.
In fact, it shocked me every time I heard from my seniors that the computational biology is not science, but merely a tool to assist, throughout my career. Some says that it is fantasy emerging from some data gimmicks, while others say it is computer/tool which is doing the analysis and not me as if pressing different keys on my keyboard throws various publication-ready results, just like a coffee vending machine. For a bioinformatics student, it will be very common that PIs from other labs will send their students to you to get their analysis done informally without giving you the due credit, merely because they think you did not put up significant effort. They forget that most bioinformaticians are biologists with no formal background of computer science. They had put up significant hard work to learn computer programming while managing their biology majors (wonder what were others upto, chanting the text from Lewin's 'Genes'?). Now since they equipped themselves with an armor of significant value, it is taken as granted by others. A strange argument was thrown at me once when another student got the things done from me and when asked whether it will be part of some manuscript, the student shouted ''it is my data, I did the hard to work to generate it, you better dont bother about its fate". This was like you donate sperm and the oocyte for IVF and then deny the hospital bill saying that it was my (well.. 'our' being heterogamous..) sperm and egg, you just fertilized it!
I even heard some reviewers accusing us for cherry-picking. Chutzpah! The cherry picking is more invasive to experimental science than to data science. The reason should be obvious. A bioinformatician/data-scientist is looking at data as a whole and supporting the hypothesis through multiple datasets and analyses, unlike an experimentalist picking a popular (to be safer) pathway or associated gene and establishing its relevance in their system through mutations/knock-downs. Interestingly, the recent trend is to look into the publicly available transcriptome or protein-protein-interaction data and pick some candidates which are likely to satisfy their hypothesis. What is it if not cherry-picking? The public availability of data and codes used/generated by bioinformaticians further endows reliability to their studies, the parallel in experimental biology is their lab notebooks, their own potty bags (for the sake of a word), with no easy way to be assessed for authenticity. It is not that I denounce the lab notebook culture completely, but rather highlight the + point of computational data centered biology.
I recalled how one PI discouraged me to work in the lab when I failed in my very first experiment. Another PI asked me what experiments I am planning and then got it done through other students. I was always asked to better focus on computation. This systematic discouragement throughout my student life left me with rather limited experimental techniques with hands on experience, despite having a craving to learn more. When I became PI, all my proposals, where I proposed experiments, were rejected citing that I have no experimental expertise. Recently, my present institution came up with an official classification of scientist as 'experimentalist' or 'theorist', leaving no room for the ones intersecting both. In fact, there is a continuous grade of scientists between experimentalist and theorist, and it is indeed a bad idea to attempt to classify people just because it comforts the policy makers. The consequences of this classification are many. They may call a meeting of all experimentalists and take a decision which will impact you if you, being theorist, pursue your interest in wet-lab through collaborations or other arrangements. They may decide upon consumable money for themselves, leaving you out. They may allot you smaller space for your students. You will be under greater academic scrutiny because you can't get the kind of excuses experimentalists cite. They may not understand the genuine problems of your end. PhD intake policies may also go against you. Being in minority, you will have the least say in the meetings, which are generally overwhelmed by the issues faced by experimentalists. The last and the most important downside is the systematic discouragement of interdisciplinary science.
Sometimes, I think that I could have pretended as an experimentalist before embarking onto my academic career and kept my computational skills as a secret armor. I also wish if my colleagues could pretend as 'scientists' more often than the elite 'expermentalists'.
PS: It is likely that my experience is unique and others did not face the same. Any text above that seems of 'generalizing' tone should be considered as error.
Image courtesy: Hodgson, "A Certain Lack of Coordination."
Image courtesy: Hodgson, "A Certain Lack of Coordination."
