Profile avatar
kristjanmoore.bsky.social
Research at deCODE genetics: genomic ancestry, ancient DNA, and whatever else needs doing. Trying to have true beliefs. ๐Ÿ‡ฌ๐Ÿ‡ง๐Ÿ‡ฎ๐Ÿ‡ธ
290 posts 183 followers 148 following
Regular Contributor
Active Commenter

Genetic architecture in Greenland is shaped by demography, structure and selection www.nature.com/articles/s41...

Nature publishes the deeply flawed "Black Death selection paper" in Oct 2022. A preprint refuting it is uploaded to bioRxiv Mar 2023. Now, 2 years later(!), Nature finally publishes the preprint as "matters arising". Number of accesses on the original paper since: 200k. Number of cites: 173. 1/2

Interesting thread on the sharing of scientific articles on Twitter vs Bluesky. For better or worse, the emigration of scientists from Twitter is real and ongoing:

Complete human recombination maps www.nature.com/articles/s41... - continuation of 25 yeas of work in building recominbation maps.

Just spent $5 to export my Twitter posts to this Bluesky account with blueark.app. Quite happy with the results. Pros: quick, easy, keeps tweet date, keeps images Cons: doesn't export tweets that quoted/replied to other accounts, so only exported 1/6 of my total posts; doesn't export poll reponses

How population stratification makes environments look like genes. A short ๐Ÿงต:

I can't help but feel that this wagon-circling and fact-massaging reflects a pattern of behaviour by the leadership of Science - of treating the science world as a kind of guild whose interests must be advanced, of explicitly leveraging the institution to promote certain political views as truth.

EXTREMELY cool and demonstrates clearly the promise of deep learning + bio. www.nature.com/articles/s41...

Great perspective on the near-term viability of human embryo editing. Much-needed nuance to the modelling exercise presented today in Nature.

In Top10 @ScienceMagazine 2024 breakthroughs of the year: Reconstructing ancient pedigrees + IBD segment sharing in aDNA. ๐Ÿ˜ฎ๐Ÿ˜Š ๐ŸŽ‰ Great write-up by @spoke32.bsky.social ! ๐Ÿ‘๐Ÿ‘

Silicon Valley has lost its mind about genetics. Huge interest in short-sleep etc variants (candidate gene nonsense), embryo selection for cognitive traits (extremely low real-life predictive ability), and Jiankui He (charlatan whose human gene editing totally failed AND made many off-target edits)

Doctors in England can record a greater number of medical conditions at inpatient visits than doctors in Wales and in Scotland. This may be why the prevalences of certain conditions as recorded in UK Biobank are surprisingly different across the countries.

I now prefer B___sky for keeping up with science. Was quiet for a while but there's now a lot of activity and discussion โ€“ probably more than what remains on here. Recommend making the jump if you haven't already! Image via @salonium includes some tools to help you start:

PhD opportunity in computational population genetics at @MPI_EVA_Leipzig: Join our fully funded project to develop IBD-segment tools, connecting high-quality genomes from 500 Black Death victims to modern Europeans. ๐ŸŒ๐Ÿ’ป๐Ÿงฌ Please spread the word! ๐Ÿ“ข๐Ÿ™Œ #aDNA #PopGen www.eva.mpg.de/de/karriere/...

"Short sleep" candidate genes are a classic example of zombie bad science which refuses to die. Announced with great acclaim 2009-2021 but decisively debunked in a 2022 biobank study (linked). Debunking has just 7 citations while originals still get 100s a year and keep going viral in popsci land.

Very excited (and a bit nervous) to announce that I will be hiring two Postdocs for my new group(!) in Copenhagen to study the Neanderthal and Denisovan DNA which survives in present-day humans. Retweet will be much appreciated :) Link for application: candidate.hr-manager.net/ApplicationI...

Heterozygous mutations in leptin receptor gene (LEPR) do not cause obesity. The finding goes against the rationale for a recent clinical trial that test setmelanotide for obesity treatment in heterozygous carriers of LEPR variants. Delplanque et al. AJHG www.sciencedirect.com/science/arti...

We have made a large update to HaploGrouper, our haplogroup assignment program. A notable addition is a fork of the ISOGG 2019 tree in which we correct the placement of 4k mutations by analysing 300k Y chromosomes. Repo in reply - see my thread on "the other place" for details: