Friday 11 December 2015

Method development

When you do method development in theoretical chemistry, you have to hide it. You cannot write a proposal about method development, you publish your new methods in technical journals, and in conference talks you have to focus on the applications. I learned in our group seminar that this is not the case in organic chemistry. As shown for example by this paper on amide-ester conversions: you can publish method developments in Nature even if it is clear that no-one except for synthetic organic chemists will ever take any interest in it. So why can't theorists do that?

Is it because theory is an irrelevant still emerging field? As a temperature check, how about we take a look at the 20 most cited papers of all time as collected by Nature:
  1. Lowry et al. "Protein measurement ..." J. Biol. Chem. 193, 265–275 (1951).
  2. Laemmli "Protein cleavage ..." Nature 227, 680–685 (1970).
  3. Bradford "Protein quantitation ..." Anal. Biochem. 72, 248–254 (1976).
  4. Sanger et al. "DNA sequencing ..." PNAS 74, 5463–5467 (1977).
  5. Chromczynski et al. "RNA isolation ..." Anal. Biochem. 162, 156–159 (1987).
  6. Towbin et al. "Protein electrophoresis" PNAS 76, 4350–4354 (1979).
  7. Lee et al. "LYP functional" Phys. Rev. B 37, 785–789 (1988).
  8. Becke "B3LYP functional" J. Chem. Phys. 98, 5648–5652 (1993).
  9. Folch et al. "Lipid isolation" J. Biol. Chem. 226, 497–509 (1957).
  10. Thompson et al. "Sequencing" Nucleic Acids Res. 22, 4673–4680 (1994).
  11. Kaplan et al. "Incomplete observations" J. Am. Stat. Assoc. 53, 457–481 (1958).
  12. Altschul et al. "Alignment search" J. Mol. Biol. 215, 403–410 (1990).
  13. Sheldrick "SHELX" Acta Crystallogr. A 64, 112–122 (2008).
  14. Altschul et al. "BLAST" Nucleic Acids Res. 25, 3389–3402 (1997).
  15. Murashige et al. "Tobacco tissue" Physiol. Plant. 15, 473–497 (1962).
  16. Perdew et al. "PBE functional" Phys. Rev. Lett. 77, 3865–3868 (1996).
  17. Folstein et al. "Mental state" J. Psychiatr. Res. 12, 189–198 (1975).
  18. Bligh et al. "Lipid extraction" Can. J. Biochem. Physiol. 37, 911–917 (1959).
  19. Southern "DNA sequences" J. Mol. Biol. 98, 503 (1975).
  20. Saitou et al. "Phylogenetic trees" Mol. Biol. Evol. 4, 406–425 (1987).
As pointed out on Curious Wavefunction, all these papers are method developments - no ideas, no discoveries. So, what's in there? Bio-stuff, the odd other topic, and ... density functionals. So, unless you are into proteins and DNA, you'd better go and develop some theoretical chemistry tools! And it is even more extreme if you look at articles published 2004-2014: Two out of the top five chemistry papers are theoretical method developments.

My point is that I should be given the opportunity to officially work on method development. So far I always had a "day job" like "excitons in DNA" or "iridium complexes" or "singlet fission" that I had to work on officially while method development was always a thing that I kind of had to hide. Or at least I had to make sure it does not divert too much attention from the application I was actually supposed to do. Everyone always tells me: If I simply write a proposal about method development (in my case wavefunction analysis tools), nobody will grant it. So why does science not have any respect for the people who provide the tools for doing science?

2 comments:

Renjith said...

I too have the same feeling and arguments whatever you wrote. Well, what I can see/infer from today's scenario is the "global climate change" happening in Scientific research field especially in "Publication Industry". Most of the journals look for bio-related/application work because this will get wider audience and thereby the journals will improve their impact factors to a high value. On the other hand if a theoretician developed an interesting and application oriented method, only very few people will be interested to read the article. This in turn will makes our life tougher. I feel this "IMPACT FACTOR" business should stop forever.

Felix said...

It is of course difficult to judge the value of new ideas. And you are right that the focus on short term "impact" may hamper some of the more exotic research that could lead to interesting results on the long run.

But my point is that electronic structure development is not exotic research: 3 of the 20 most cited papers of all fields are in that area. And that makes it even stranger that it is still kind of a marginalized field.

but actually I am happy with my job and should not complain ... :)