Who is at #AGBTGM26 ?
23.02.2026 22:11 β π 0 π 0 π¬ 0 π 0Who is at #AGBTGM26 ?
23.02.2026 22:11 β π 0 π 0 π¬ 0 π 0Science Bluesky is now much better than Science Twitter isβ¦ but I still miss peak pre-pandemic Science Twitter.
14.02.2026 19:04 β π 16 π 1 π¬ 2 π 0Fun news! @gcbias.bsky.social and I are teaching a 2-week online population genetics workshop this summer to raise money for the Center for Population Biology at UC Davis. We're trying to gauge interest -- please fill this out if you might be interested! And please share broadly!
09.02.2026 18:24 β π 137 π 169 π¬ 7 π 5I mean, death from infectious disease clearly has a genetic component related to the immune systemβ¦ the distinction between intrinsic and extrinsic seems a little slippery.
31.01.2026 18:25 β π 3 π 2 π¬ 0 π 0Clever evolutionary geneticists are liking my post, so I guess I might be on to somethingβ¦
31.01.2026 01:13 β π 1 π 0 π¬ 0 π 0
I haven't thought deeply about this, but ... if heritability is the proportion of the variance attributable to genetics, and I remove some variance attributable to "the environment", then should I be surprised that the heritability goes up? @sashagusevposts.bsky.social help!?
doi.org/10.1126/scie...
We're hiring! We're seeking a talented statistical geneticist to join the team. Come work with the greatest genetic dataset in the world, and help shape the 23andMe Research Institute in our nonprofit era!
23andme.wd5.myworkdayjobs.com/en-US/23/det...
Brenna Henn Wanted to Improve Genetic Medicine. Then Her N.I.H. Grant Was Cut.
02.01.2026 21:16 β π 39 π 19 π¬ 3 π 2
I've been thinking about the "virtual cell" concept and wanted to write up a few thoughts. Specifically on how I think the prior experience in GWAS informs the most likely way these models will be useful.
andrewcarroll.github.io/2025/12/23/t...
Completely agree. I was lucky to get to work with Richard on the 1000 Genomes. He was always so generous with ideas and insights.
02.12.2025 18:34 β π 0 π 0 π¬ 0 π 0Congratulations to Richard Durbin on being awarded our Genetics Society Medal!
14.11.2025 12:40 β π 83 π 26 π¬ 1 π 3
Keeping fuel duty at the same level for 15 years, while phasing in a cumbersome-to-implement tax on EV owners seems insane.
www.bbc.com/news/article...
PRSformer: Disease Prediction from Million-Scale Individual Genotypes
Figure 1
Figure 2
Figure 3
PRSformer: Disease Prediction from Million-Scale Individual Genotypes [new]
Deep learning w/ neighborhood attention predicts disease from million-scale genotype data by learning genetic interactions.
I wrote a little bit about the "missing heritability" question and several recent studies that have brought it to a close. A short π§΅
21.11.2025 22:33 β π 351 π 168 π¬ 14 π 22The field has also gained leverage from personal genomics initiatives, as in this study of ~52k adults answering βyesβ to βhave you been diagnosed with dyslexia?β vs >1 million responding βnoβ, which found 42 associated loci + high genetic overlap with directly assessed reading/language skills. 10/n
18.11.2025 09:59 β π 4 π 2 π¬ 1 π 0
Preprint: www.biorxiv.org/content/10.1...
Github: github.com/23andMe/PRSf...
Feedback welcome!
Huge congrats to the team: Payam Dibaeinia, Chris German, Suyash Shringarpure, and Aly Khan on getting this out the door. Come by Poster Session 1 on Wed 3 Dec 11 a.m. - 2 p.m PST at #NeurIPS2025 San Diego!
28.10.2025 22:23 β π 0 π 0 π¬ 1 π 0
We show that non-linear gains emerge for immune-
mediated diseases at N > 1M. Harnessing phenome-wide (NxGxD) models seems a fruitful direction to borrow information across traits.
Core design: Neighborhood attention (O(L) complexity/layer) models the genome in ~100kb intervals, akin to LD blocks. By stacking layers, the model learns local dependencies first, then integrates long-range info and makes predictions feasible at genome scale.
28.10.2025 22:23 β π 1 π 0 π¬ 1 π 0We trained PRSformer at 23andMe Research Institute on population scale data (N > 1M persons; G > 100k variants; D > 10 autoimmune/inflammatory traits). PRSformer significantly outperforms linear baselines and summary-statistic methods (LDPred2) derived from the same cohort.
28.10.2025 22:23 β π 0 π 0 π¬ 1 π 0
Delighted to see our method, PRSformer, at #NeurIPS2025! PRSformer is AI model for population-scale disease-risk prediction from individual genomes. It lays the groundwork for phenome-wide risk prediction.
www.biorxiv.org/content/10.1...
PRSformer: Disease Prediction from Million-Scale Individual Genotypes https://www.biorxiv.org/content/10.1101/2025.10.26.684578v1
27.10.2025 18:32 β π 2 π 2 π¬ 0 π 0The brilliant Sarah Senk, together with Taiyo Inoue, has just launched a podcast that explores the implications of AI for higher education: My Robot Teacher. Please give it a listen! #MyRobotTeacher #HigherEd www.youtube.com/watch?v=H2Ta...
02.07.2025 19:50 β π 2 π 0 π¬ 0 π 0YOUR 2024/25 CARABAO CUP WINNERS π
16.03.2025 20:09 β π 1042 π 212 π¬ 41 π 40On the x-axis is every human gene, ranked by number of publications containing mention of the gene name. Lots left to discover...
15.02.2025 15:57 β π 4 π 2 π¬ 1 π 0
We're hiring a postdoc to help shape our autism research program; please consider applying.
www.23andme.com/careers/jobs...
"the science of extreme longevity continues as an immense joke."
www.nytimes.com/2025/01/20/o...
We're looking for a talented statistical geneticist to come work with us!
www.23andme.com/careers/jobs...
We're looking for a talented statistical geneticist to come work with us!
www.23andme.com/careers/jobs...
Humans tend to inherently believe that context matters. But context doesnβt seem to matter all that much in genetics.
Epistasis between mutations doesnβt seem to influence their stability in this amazing saturation mutagenesis paper.
www.nature.com/articles/s41...