Surag Nair
@suragnair.bsky.social
Machine learning and genetics @Genentech. Previously CS PhD @Stanford. suragnair.github.io
Agents for comp bio are advancing rapidly, but evals are lagging. Current benchmarks can be overly prescriptive. Full analysis vignettes are hard to verify. We introduce CompBioBench: 100 diverse, challenging, verifiable tasks. We benchmark Codex and Claude Code. biorxiv.org/content/10.6... 1/9
We are hiring a PhD intern for Summer 2026 in ML for regulatory genomics at ReLU/BRAID/Genentech! Work on DNA sequence models for the noncoding genome (e.g. DNA design, models of MPRA and genetic variants)! 🥳
2026 Summer Intern - Biology Research | AI Development in South San Francisco, California, United States of America | Students & Graduates at Genentech
Apply for 2026 Summer Intern - Biology Research | AI Development job with Genentech in South San Francisco, California, United States of America. Students & Graduates at Genentech
lnkd.in
An artistic reinterpretation of the Nona model's schematics
Introducing Nona! 🧬 @suragnair.bsky.social 's brilliant idea to unify siloed genomic AI. Nona learns jointly from DNA seq + functional data, enabling new ways of modeling genomic data!
Surag Nair@suragnair.bsky.social · 10mo agoExcited to share Nona: a unifying multimodal masking framework for functional genomics. Models for DNA have evolved along separate paths: sequence-to-function (AlphaGenome), language models (Evo2), and generative models (DDSM). Can these be unified under a single paradigm? 1/15
Excited to share Nona: a unifying multimodal masking framework for functional genomics. Models for DNA have evolved along separate paths: sequence-to-function (AlphaGenome), language models (Evo2), and generative models (DDSM). Can these be unified under a single paradigm? 1/15
[SAVE THE DATE] MLCB 2025 is happening Sept 10-11 at the NY Genome Center in NYC! Attend the premier conference at the intersection of ML & Bio, share your research and make lasting connections! Submission deadline: June 1 More details: mlcb.github.io Help spread the word—please RT! #MLCB2025
We are hiring an intern to work with our team at Genentech next summer, on exciting projects related to deep learning for DNA/RNA sequences. Please share and apply! roche.wd3.myworkdayjobs.com/ROG-A2O-GENE...
2025 Summer Intern - Biology Research | AI Development
2025 Summer Intern - Biology Research | AI Development Department Summary At Genentech Research & Early Development (gRED) we have initiated an exciting journey to bring together and further stren...
roche.wd3.myworkdayjobs.com
An interesting diagnostic application of CRISPR is to activate expression of genes in tissues where they are not normally expressed. This is useful when studying functional consequence of suspect pathogenic variants in genes that are restricted to inaccessible tissues like brain, eyes etc. 1/
Gene regulation involves thousands of proteins that bind DNA, yet comprehensively mapping these is challenging. Our paper in Nature Genetics describes ChIP-DIP, a method for genome-wide mapping of hundreds of DNA-protein interactions in a single experiment. www.nature.com/articles/s41...
ChIP-DIP maps binding of hundreds of proteins to DNA simultaneously and identifies diverse gene regulatory elements - Nature Genetics
ChIP-DIP (ChIP done in parallel) is a highly multiplex assay for protein–DNA binding, scalable to hundreds of proteins including modified histones, chromatin regulators and transcription factors, offe...
nature.com
Excited to share our latest preprint on scE2G – a new model to link enhancers to target genes using single-cell data – with state-of-the-art performance across multiple perturbation benchmarks. biorxiv.org/cgi/content/... Read more below! 1/12
Mapping enhancer-gene regulatory interactions from single-cell data
Mapping enhancers and their target genes in specific cell types is crucial for understanding gene regulation and human disease genetics. However, accurately predicting enhancer-gene regulatory interac...
biorxiv.org
By demand, I've created the final starter pack in my ML Personality Starter Pack Series. I'm uncertain who belongs in this starter pack and so if you think you better fit in the Grumpy ML or Unreasonably Upbeat ML starter packs, let me know. (Self) nominations welcome go.bsky.app/5Suyk58
A reminder for new folks. BlueSky does not have an algorithm to raise posts of interest. It is incumbent upon you to do so via reposting. Liking things only provides feedback to the poster but does not promote the post. The signal to noise ratio is taking a brief beating, so great to help curate.
1/ Introducing SCimilarity, a new foundation model to explore single-cell RNA-seq data across tissues and diseases! It learns a common measure of cell similarity by training a deep metric learning model on millions of cells from various human tissues and conditions. www.nature.com/articles/s41...
A cell atlas foundation model for scalable search of similar human cells - Nature
Nature - A cell atlas foundation model for scalable search of similar human cells
nature.com
Impressive study from Alex Stark lab identifying 3 new types of silencer elements in Drosophila New TF binds a particular isolated motif in non-accessible sites to recruit G9a. Deletion of these elements can lead to upregulation of nearby gene www.sciencedirect.com/science/arti...
We also review variant effect prediction evaluations that have been performed to date on genomic deep learning models, highlighting strengths and limitations of current models and the need for more comprehensive evaluation. 3/4
Super excited to share our review on genomic deep learning models for non-coding variant effect prediction, with Ayesha Bajwa and Nilah Ioannidis. We’d like this review to be a useful resource, and welcome any feedback, comments, or questions! 1/4 arxiv.org/abs/2411.11158
Leveraging genomic deep learning models for non-coding variant effect prediction
The majority of genetic variants identified in genome-wide association studies of complex traits are non-coding, and characterizing their function remains an important challenge in human genetics. Gen...
arxiv.org
Really intriguing model of AP-1 driven aging. @anshulkundaje.bsky.social @suragnair.bsky.social - thinking of your AP1-related results in the ChromBPNet paper here...
Christian Nefzger@nefzgerlab.bsky.social · 2y ago8/9 🧵 Our study indicates that AP-1–linked chromatin opening drives organismal maturation by disrupting the activity of cell identity TFBS-rich early-life REs, thereby progressively shutting down developmental processes to reprogram the transcriptome towards adult tissue function.
Highly recommend if you’re considering a PhD in ML for genomics. Jacob is a phenomenal scientist, knows how to dive deep into the nitty-gritty of things, is incredibly patient, and just a very fun person to work with. His weakness? Incosistent joke quality.
Jacob Schreiber@jmschreiber91.bsky.social · 2y agoMy goal is to understand the regulatory role of every nucleotide in the genome, and how this changes across every cell in the human body. If you are interested in doing a Ph.D. with me at UMass Chan Medical (Genomics and Comp Bio Department), see the links below. Deadline is Dec 1st.
Two BioML starter packs now: Pack 1: go.bsky.app/2VWBcCd Pack 2: go.bsky.app/Bw84Hmc DM if you want to be included (or nominate people who should be!)
Kevin K. Yang 楊凱筌@kevinkaichuang.bsky.social · 2y agoI tried to make a bioml starter pack. DM if you want me to add or remove you? go.bsky.app/2VWBcCd
Dusting off this account since it seems like BlueSky is suddenly the place to be! Here's a couple of resources for anyone interested in gene regulation, chromatin, or genomics: First, a long list of relevant accounts: bsky.app/profile/shau... (will make some starter packs at some point)
Incredible resource for DNA binding protein specificity from the Codebook consortium. www.biorxiv.org/content/10.1... www.biorxiv.org/content/10.1... www.biorxiv.org/content/10.1... www.biorxiv.org/content/10.1... Am reading these in detail & will have a lot to say. Stay tuned for thoughts.
Extensive binding of uncharacterized human transcription factors to genomic dark matter
Most of the human genome is thought to be non-functional, and includes large segments often referred to as "dark matter" DNA. The genome also encodes hundreds of putative and poorly characterized tran...
biorxiv.org