1/ Exactly the same theory was formalized by Russel Lande in 1976 and developed further by other evolutionary quantitative geneticists, and now the molecular biologists and medical scientists are re-discovering it in 2018 as "the omnigenic model" #ReinventingTheWheel
2/ The historical illitteracy and underappreciation of quantitative genetics as a tool from both molecular biologists and unfortunately also some molecularly oriented evolutionary biologists never stops to amaze me
3/ About ten years ago, when I discussed how to interpret and solve "the missing heritability problem" with a molecular ecology postdoc, I got the following (naive) suggestion: "Why not sequence more?"
4/ My answer: Think more, sequence less. Or think more before you sequence. We need better statistical tools, better models and increased appreciation of the implications of the infinitisemal model and the power of quantitative genetics.
5/ Also, I recently attended a nice talk demonstrating phenotypic selection on quantitative traits, and I was amazed by the speaker who said at the end of the talk that the next step is "to find candidate genes for the trait".
6/ But how exactly would our understanding of selection on a quantitative trait increase by finding some candidate gene(-s) that explain 5-10 % of the phenotypic variation (at most)?
7/ Don't get me wrong: I am using molecular tools myself in my research lab and they are often extremely valuable. But one should think hard if, why and when to use them, given the costs. Too many evolutionary ecologists feel pressure to use these tools to show they are "modern"
8/ For many selection studies in the wild that focus on quantitative traits with highly polygenic architectures and traits governed by many loci of small effects, it is perfectly legitimate and valuable to not use any molecular tools at all
9/ In short: studying phenotypic evolution and natural selection in the wild is a noble goal in itself, as it has been in the genomic era and will continue to be in the postgenomic era
10/ As evolutionary biologists interested in phenotypes, I think we should not fall in to the reductionist trap, but be pluralistic and accept that neither "top-down" (phenotypes) or "bottom-up" (molecules) approaches are sufficient, but complementary to each other.
