Today in Science Magazine, we report a new DNA editing technology to seamlessly write massive changes into the right place in the human genome. The reason gene editing hasn't transformed human health is that current gene editing technologies like CRISPR are very limited. The

Do you know ~60% of human SVs fall in ~1% of GRCh38? See our new preprint: arxiv.org/abs/2509.23057 and the companion blog post on how we started this project and longdust: lh3.github.io/2025/09/29/on-…. Work with Qian Alvin Qin

A high-coverage genome from a 200,000-year-old Denisovan 💀 biorxiv.org/content/10.110… 🧬🖥️🧪

"Systematic artifacts from Illumina two-color chemistry confound variant identification and actionability in clinical panels" medrxiv.org/content/10.110…

Stop the hype. IQ embryo selection is not about "better babies"; it's a snake oil scam fueled by bad science. It has zero control over the thousands of linked disease phenotypes. This is reckless genetic tinkering, not public, not healthy.

Ah yes, the classic “students only like me when I stop making them miserable” revelation. It’s almost like teaching is supposed to help people learn instead of testing who can survive your ego-driven obstacle course.

Required reading for cell biologists to get a sense of basic statistical principles! nature.com/articles/s4155…

Enjoyed reading the recent paper from gnomAD team on the modifiers of penetrance of clinically-relevant variants in 800k individuals in the gnomAD database. One example highlighted in the paper is extremely fascinating! It's a beautiful example of a loss of function (LOF)

Repair of DNA double-strand breaks leaves heritable impairment to genome function | Science science.org/doi/10.1126/sc…

Thanks for your deep investigation 四川麻辣燙 where is the investors due diligence?

579 high-quality human genomes from Human Pangenome Reference Consortium, Arab Pangenome and individual papers (CHM13, CN1, KSA001, I002C, YAO and KOREF1). Sequences available in the AGC format (3.7GB) and FM-index in the ropebwt3 format (20.3GB). For details, see github.com/lh3/human-asm

Modern GWAS can identify 1000s of significant hits but it can be hard to turn this into biological insight. I'm excited to share our new work combining genetic associations and Perturb-seq to build interpretable causal graphs, out today in nature:

The scikit-bio paper in online in Nature Methods! Many thanks to our collaborators, community contributors and reviewers! We couldn’t have done it without you. nature.com/articles/s4159… #Bioinformatics #OpenSource

While the rise the ML "fit anything" approach is certainly powerful, I have noticed an accompanying decline in some quantitative reasoning skills, stuff like the right and wrong ways to normalize, etc. Maybe it doesn't matter anymore, but it feels like something has been lost.

Evaluating the effects of archaic protein-altering variants in living human adults science.org/doi/10.1126/sc… 🧬🖥️🧪

What sets the mutation rate of a cell type in an animal species? biorxiv.org/content/10.648… 🧬🖥️🧪

We are there! Our panelada team is growing! :) Time to weight and update the (hyper)parameters!

This is probably the best essay I’ve read on the subject of mathematics in biology. (Gunawardena 2014)

I just updated the scDiffEq Python API (v1.1.0) to include animated visualizations of simulations. You can generate GIFs (or still images) showing: - Temporal gene expression dynamics - UMAP projections of simulated trajectories During my PhD, I found animations like these

How does AlphaFold select protein conformations? Its transformer model (Evoformer) associates them with sparse sequence patterns. We developed a method to identify these patterns readily🧵 biorxiv.org/content/10.648…