When and how to “go for the genes”


A new special issue of Molecular Ecology, entitled “Detecting selection in natural populations: making sense of genome scans and towards alternative solutions”, is coming down the line, and a few articles from that issue are starting to appear as newly-accepted.

Seeing those words made me think back to some papers I downloaded a while back based on an interesting tweet:

Evolution is a fundamentally genetic process, but the reciprocal connections between alleles, phenotypes, and environment are not fundamentally tractable. All three of these papers are various stances on the relationship between identifying genes that underlie traits of interest and understanding the evolution of those traits. The second link is a strong critique of the genotype-phenotype map pursuit and suggests that chasing underlying genes and alleles is distracting evolutionary biologists from the study of phenotypes. The third link is from a 2011 paper that covers the problem of “missing heritability”, an issue elegantly summarized by The Molecular Ecologist previously.

The first link takes you to a new paper appearing in the most recent issue of Evolution and provides yet another conversation starter by suggesting a blueprint for when “going for the genes” may be necessary and when it’s not. Rausher and Delph lay out a total of ten study objectives, broken up into two categories.

A. When explanations of evolutionary processes could be made without identifying the genes responsible for the phenotype of interest:

  1. Explaining evolutionary change and divergence in quantitative traits
  2. Detecting tradeoffs
  3. Explaining evolutionary change and divergence in Mendelian traits

B. When gene identification in justified

  1. Any study of molecular evolution
  2. Drift vs. selection
  3. Analysis of parallel evolution
  4. Understanding asymmetries in evolutionary transition rates
  5. Evaluating the cause of the trait-trait and trait-fitness correlations
  6. Costs of adaptation
  7. Selection-component analysis in undisturbed natural populations

I admired the reasonable and constructive dialog here, including many “how-to” sections to back up their claims, which are especially helpful when you are someone looking from the outside-in at this “debate”.

Or, if you’re feeling feisty and find your work in category A, here is a potential script for your next interaction with a reviewer or conference talk attendee:

Many evolutionary biologists have been asked at one time or another why they haven’t tried to determine what genes underlie the evolutionary issues they examine. In our view, a legitimate answer is that doing so would not significantly enhance our understanding of those issues.

but wait!

Given the vast range of issues knowledge of the relevant genes can address, however, in providing that answer, there are situations in which one must be able to articulate why discovering the genes is indeed irrelevant.

“Have an opinion that you can support” continues to be timeless advice, whether you are “going for the genes” or not.


Rausher, M. D., & Delph, L. F. (2015). When does understanding phenotypic evolution require identification of the underlying genes?. Evolution, 69(7): 1655-1664.

Rockman, M. V. (2012). The QTN program and the alleles that matter for evolution: all that’s gold does not glitter. Evolution, 66(1), 1-17.

Travisano, M., & Shaw, R. G. (2013). Lost in the map. Evolution, 67(2), 305-314.


About Rob Denton

I'm an Assistant Professor in the Division of Science and Math at the University of Minnesota Morris. I'm most interested in understanding the evolutionary/ecological consequences of strange reproduction in salamanders (unisexual Ambystoma). Topics I'm likely to write about: population and landscape genetics, mitonuclear interactions, polyploidy, and reptiles/amphibians.
This entry was posted in adaptation, association genetics, Molecular Ecology, the journal and tagged , , . Bookmark the permalink.