Jeffrey Chang
08 April 2015
Biological data will continue to pile up unless those who analyse it are recognized as creative collaborators in need of career paths, says Jeffrey Chang.
The article makes very salient and relevant points - especially the section called "Routinely unique". That is what we see all the time and makes bioinformatics a much more difficult job than what one might expect.
A standard RNA-seq analysis is perhaps a 15 minute job to set up and launch on a compute node - will finish in less than a day. But if the system is even a bit outside the most common we might just end up with a two week long "slugfest".
What the scientific community should understand is that, "Methods" papers are generally highly cited than "biology" papers (The top-100 papers). The "Methods" papers could even be expanded to some papers describing experimental techniques (like PCR or Sanger-sequencing) and not necessarily have any biological discovery. So its really important that expertise of bioinformaticians be appropriately recognized as bioinformatics is in the heart of every biological research.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
The article stated that most "core" bioinformaticians spend maybe 5% of their time writing these highly cited methods papers. So this is not a solution to the problem posed by the Nature commentary. I also disagree with the central claim as well: "Relatively little effort is needed to help train, recognize, reward and retain applied bioinformaticians. Physics, with its long history of massive, highly integrative projects has begun to sort out career paths, authorship norms and due credit for those in purely collaborative roles — most notably in beam-line service provision."
The solution is not as simple as that, at least not within university context (although I must state that I am not familiar with "beam-line service provision"). There simply is no way to have Ph.D. trained scientists to be happy to carry out service roles in the long term. Also, I don't think that the ratio of beam-line technicians to researchers is nearly as high as the ratio of bioinformaticians to biologists needs to be. And I doubt that the beam-line specialists on the whole know more about statistics and mathematics than the particle physicists. The expertise that "biologists" and "bioinformaticians" have is genuinely complimentary and non-overlapping at a much deeper level than that.
The obvious solution, that is not really explored, is to have bioinformaticians as second or third "first authors" and to preferentially fund and carry out research where the bioinformatics component is better integrated, not just simple "profiling" studies. It would be also easier to recruit bioinformaticians to carry out analysis that they have a hand in designing.
Additionally (as the article states) "biologists" need to learn to analyze their own studies. This is already happening but training in this regard is severely lacking. Anyway, more and more bioinformaticians are striking it on their own and "Data Analysis" papers are becoming more common also in high-level journals. This is maybe where, and how, bioinformaticians make their most notable contributions.