In likely the largest deal ever for a seed-stage biotech/techbio company, DCVC-backed Relation and GSK will seek to create a new gen of drugs for fibrotic disease and osteoarthritis. Big congrats to both companies! More from DCVC GP Jason Pontin 👇 bit.ly/3ZunHo3

Tabula Sapiens 2.0, a reference atlas with >1.1 million cells from 28 human tissues, reveals insights into transcription factor expression, cell senescence, & sex-specific gene expression. Explore the findings: biorxiv.org/content/10.110…

1/ Excited to see moscot-tools.org published in nature! We scaled Optimal Transport (OT) in single-cell genomics & added multimodality together with spatiotemporal trajectory inference, finding exciting new biology in the pancreas! 🚀 Read at nature.com/articles/s4158…

Super excited to share our new study from the Jason Buenrostro Lab in nature! We developed a computational method for tracking transcription factor and nucleosome binding using single-cell ATAC-seq and deep learning. Paper: nature.com/articles/s4158…

New preprint -- arxiv.org/abs/2501.19178 Old school math bio treatment of perturbation prediction! #CRISPR #singlecell CC - Relation Lindsay Edwards Dr. Caroline Uhler Hagen Triendl Thomas Gaudelet

🙌🙌🙌 nature.com/articles/s4158…

Will there be a singular "DeepSeek moment" for the burgeoning intersection of ML + Biology? I answer this and more in my newest article in the Timmerman Report. A Brief 🧵 timmermanreport.com/2025/02/deepse…

Read the scTrends.org take on the AI arms race to generate single-cell data sctrends.org/p/the-single-c… Thanks Adam Cribbs !

Love seeing these new models, but computational predictions need real lab validation. We need community-wide DNA model benchmarking for generative processes! Real-world function > theoretical performance 🧬🔨

Published on bioRxiv - the world's first 100M single cell dataset! Tahoe-100M includes 52,886 unique drug perturbations and was run through Parse’s GigaLab in < 1 month!

How can causal machine learning help answer causal questions in single-cell genomics—like how genes interact and influence phenotypes? Our latest Nature Genetics paper explores this challenge, along with approaches to generalizability, interpretability, and modeling cell dynamics.

I'm thrilled to share our new #bioRxiv preprint from ⁦Junyue⁩ lab at ⁦Rockefeller University⁩! We present PerturbFate, a high-throughput single-cell CRISPR platform profiling nascent/pre-existing RNA, steady-state RNA & chromatin accessibility. biorxiv.org/content/10.110…

Just released today in Nature Methods. If you haven't tried Proseg's cell segmentation method on your spatial transcriptomic data yet, maybe you should. See how it stacks up to other methods with data from multiple platforms. nature.com/articles/s4159…

Solving the "replication crisis" is dumb because (1) it's not a crisis, (2) the medicine is worse than the disease. Much of this is because people want science to be more like engineering, and it's not. It's fundamentally different and trying to treat engineering processes like

Cell-type deconvolution methods for spatial transcriptomics | Nature Reviews Genetics nature.com/articles/s4157…

Interested in #singlecell RNA sequencing? Don’t miss this talk by Charlie Roco (CTO and Co-Founder, Parse Biosciences) at The Festival of Genomics & Biodata, Boston. Register now: hubs.la/Q03qRG800 #FOGBoston

Most biological processes, like stem cell differentiation, branch into multiple fates, but current trajectory inference methods only predict single paths.🌳To fix this, we introduce BranchSBM 🌿, from our unstoppable Sophia Tang ! 🌟 📜: arxiv.org/abs/2506.09007 💻:

Limitations of cell embedding metrics assessed using drifting islands | ⁦Nature Biotechnology⁩ ⁦Jure Leskovec⁩ ⁦Hanchen Wang⁩ Aviv Regev nature.com/articles/s4158…