Dear diary,

Time for a recap of the On unicorns and genes blogging year. During 2019, your friendly neighbourhood genetics blog mostly kept to its schedule of four posts per month with some blog vacation in summer and in December.

This added up to a total of 43 posts (including this one), only one of them in Swedish (Gener påverkar ditt och datt, reflecting on how genome-wide association is often reported as if it was something else), and three of them posts about three first-author papers that came out in 2019:

Paper: ”Integrating selection mapping with genetic mapping and functional genomics”
Paper: ”Sequence variation, evolutionary constraint, and selection at the CD163 gene in pigs”
Paper: ”Removal of alleles by genome editing (RAGE) against deleterious load”

Now, let’s pick one post per month to represent the blogging year of 2019:

January: Showing a difference in means between two groups. This is one of those hard easy problems: Imagine we have an experiment comparing the means of two groups; how do we show both the data and the estimate of the difference, with uncertainty, in the same plot? In this little post, I try a more and a less radical version. I’ve since used the less radical version a couple of times.

February: ”We have reached peak gene and passed it”. Comment on an opinion piece that argued that revisions to the gene concept have important implications for modern genetics, and people need to be told. I agreed about a lot of the criticisms, but thought they should have nothing to do with gene concepts.

March: Journal club of one: ”Biological relevance of computationally predicted pathogenicity of noncoding variants”. Journal club post about a then recent paper about how variant effect prediction, especially for noncoding variants, is really hard, and not easy to evaluate fairly either.

April: Greek in biology. Comment on a stimulating essay about multilingualism in biology.

May: What single step does with relationship. A simulation and a couple of heatmaps to try to understand how the single step method describes relationship between individuals by blending genomic and pedigree relatedness.

June: Simulating genetic data with R: an example with deleterious variants Post from my talk at the Edinburgh R users group.

July: Using R: Correlation heatmap with ggplot2. I thought I’d update one of my most viewed posts, which was becoming too embarrassingly outdated. Unfortunately, I also broke its old address, and that wasn’t so smart.

August: Blog vacation.

September: Genes do not form networks. They just don’t. It seems DiFrisco & Jaeger (2019) don’t think so either. I would like to return to this later.

October: Using R: Animal model with simulated data. What it says on the tin: example of simulating a pedigree with AlphaSimR and fitting an animal model.

November: Using R: from gather to pivot. Introduction to, and celebration of, one of the best changes in the tidyverse.

December: Interpreting genome scans, with wisdom. Opinions about genome-wide association, precipitated by Eric Faumann’s Twitter account.

This is also the year I moved from the Roslin in Scotland to Uppsala, Sweden, for the second phase of the mobility grant-supported postdoc. Here are many of my worldly possessions, being hauled:

What will happen on here in 2020? The ambition is to keep a reasonably regular schedule, post about papers as they come out, and maybe write some more in Swedish.

I’ll also have a blog anniversary. The blog, admittedly rather different the first years, started in earnest in June 2010. I’m not sure how to celebrate, but I feel like I should follow up on one of the first posts (about linkage mapping and the probably spurious ”Gay Gene” at Xq28), now that Ganna et al. (2019) put genetic mapping of sexual orientation is in the news again.