When is on-person evolution dominated by neutral vs adaptive forces? If you're looking for a postdoc position and this is a question that interests you, please see my lab website and reach out! [More info in next posts]

We studied the evolution of genome-wide linkage disequilibrium and its influence on genetic architecture with genetic drift. We found that the Bulmer effect is population size and demography dependent. For nerds: genetic architecture follows from solving PDEs.

I’m looking to hire an NIH funded postdoc to join our lab here at Davis (please RT). Ongoing areas of interest include: 1) polygenic signals of selection from time series, ARGs, GWAS, etc 2) The interpretation of GWAS, polygenic scores, & sources of confounding

Ah, I see — that makes sense. In that case I’d keep “与数据据理力争”. My main suggestion was really the first half: “心中有‘树’” feels like a more obvious pun to me than “按图溯祖”. I don’t think my rewrite actually improves the second half much.

Maybe you should take the pun all the way: 心中有“树”，让数据“据”理力争. (Now it's tree not graph tho)

Update: The academic flight from Twitter is now complete and appears permanent. April 2023, after the pandemic research surge, we recorded 2,234,307 posts sharing research on Twitter. As of April this year, "X" had 896,741 posts about research. That's a 60% drop in traffic.

FGM seems to be a particular formalization of an adaptive landscape: it assumes a multidimensional phenotypic space with a single fitness optimum. Adaptive lanscape is more abstract and conceptual typically with many adaptive peaks and, importantly, it can be itself evolvable.

This was a great collaboration with @kgslab.bsky.social and @evodaveo.bsky.social. Many thanks also to @wormsrock.bsky.social and @lmchev.bsky.social for feedback on an early draft, and to the reviewers for constructive feedback that really helped sharpen the framing of the manuscript.

A high rD may also be biologically informative: it could reflect trait redundancy, developmental coupling, or geometric coupling, where many genetic perturbations affect the two traits in similar ways.

In yeast morphology, trait pairs with similar rG could have very different rD, and in some cases a small number of genomic regions strongly shaped the observed correlation. So the value of a genetic correlation alone can miss a lot of its underlying structure.

To separate these possibilities, we developed a marker-effect framework that distinguishes candidate large covariance contributors from the polygenic background. We summarize the latter with rD: how consistently small/moderate marker effects align across the genome.

The basic idea: the same genetic correlation (rG) can be produced by many loci with aligned pleiotropic effects, or by a few loci making unusually large contributions to covariance. Those architectures may look similar at the level of rG, but need not have the same biological basis or stability.