Tessera
Folgen

Peter Koo

@pkoo562.bsky.social

AI4Science researcher. Associate Professor @CSHL. My lab advances AI for genomics and healthcare! http://koo-lab.github.io

130 Posts3215 Follower1272 Folgt
Peter Koo@pkoo562.bsky.social · 3mo ago

Hello Bluesky! Virtual registration is open for the 90th CSHL Symposium: AI in Biology! Amazing lineup of speakers! Join live on Zoom w/ Q&A, and access recordings on demand for ~48 hrs after each talk, with archive access pending speaker approval. meetings.cshl.edu/virtualregis... #cshlsymp26

CSHL Virtual Meetings Registration Form

Cold Spring Harbor Laboratory Meetings & Courses -- a private, non-profit institution with research programs in cancer, neuroscience, plant biology, genomics, bioinformatics.

meetings.cshl.edu

Peter Koo@pkoo562.bsky.social · 7mo ago

Save the date: The premier AI x Bio meeting of 2026 will be held at CSHL on May 26-31! The program brings together 50+ invited leaders in genomics, transcriptomics, protein design, drug discovery, neuro-AI, pathology, agentic AI, and more! Abstract due: March 26 meetings.cshl.edu/meetings.asp...

90th Symposium: AI in Biology

Cold Spring Harbor Laboratory Meetings & Courses -- a private, non-profit institution with research programs in cancer, neuroscience, plant biology, genomics, bioinformatics.

meetings.cshl.edu

Peter Koo@pkoo562.bsky.social · 10mo ago

Which mutations rewire function of regulatory DNA? Excited to share SEAM: Systematic Explanation of Attribtuion-based Mechanisms. SEAM is an explainable AI method that dissects cis-regulatory mechanisms learned by seq2fun genomic deep learning models. Led by @EESetiz 1/N 🧵👇

Bild
Peter Koo@pkoo562.bsky.social · 11mo ago

Check out a Research Highlights on our work at @naturemethods by Lin Tang! www.nature.com/articles/s41...

Benchmarking genomic language models - Nature Methods

Nature Methods - Benchmarking genomic language models

nature.com

Peter Koo@pkoo562.bsky.social · last yr.

Our work on "Evaluating the representational power of pre-trained DNA language models for regulatory genomics" led by @AmberZqt with help from @NiraliSomia & @stevenyuyy is finally published in Genome Biology! Check it out! genomebiology.biomedcentral.com/articles/10....

Peter Koo@pkoo562.bsky.social · last yr.

*Easter egg alert* NOT in the published paper. We also benchmarked Evo 2 and while it did better than other gLMs (consistent that scale can improve gLMs), it still falls short of a basic CNN trained using one-hot sequences and far short of supervised SOTA

Peter Koo@pkoo562.bsky.social · last yr.

Our work on "Evaluating the representational power of pre-trained DNA language models for regulatory genomics" led by @AmberZqt with help from @NiraliSomia & @stevenyuyy is finally published in Genome Biology! Check it out! genomebiology.biomedcentral.com/articles/10....

Peter Koo@pkoo562.bsky.social · last yr.

Our work on "Evaluating the representational power of pre-trained DNA language models for regulatory genomics" led by @AmberZqt with help from @NiraliSomia & @stevenyuyy is finally published in Genome Biology! Check it out! genomebiology.biomedcentral.com/articles/10....

Evaluating the representational power of pre-trained DNA language models for regulatory genomics - Genome Biology

Background The emergence of genomic language models (gLMs) offers an unsupervised approach to learning a wide diversity of cis-regulatory patterns in the non-coding genome without requiring labels of ...

genomebiology.biomedcentral.com

Anshul Kundaje@anshulkundaje.bsky.social · last yr.

One thing that really bothers me with the new "virtual cell" terminology is that it is currently largely focused on a very narrow definition of models that can predict effects of trans perturbations (gene dosage, drugs etc) on gene expression. 1/

Anshul Kundaje@anshulkundaje.bsky.social · last yr.

This a really exciting leap forward for genomic sequence to activity gene regulation models. It is a genuine improvement over pretty much all SOTA models spanning a wide range of regulatory, transcriptional and post-transcriptional processes. 1/

Žiga Avsec@avsecz.bsky.social · last yr.

Excited to launch our AlphaGenome API goo.gle/3ZPUeFX along with the preprint goo.gle/45AkUyc describing and evaluating our latest DNA sequence model powering the API. Looking forward to seeing how scientists use it! @googledeepmind

Peter Koo@pkoo562.bsky.social · last yr.

Building virtual cells is a great goal in the age of AI, but it requires far more than training transformers with scRNAseq. *Scaling* as the primary strategy with hopes of emergent properties is lazy. Will the plan to fuse representations across mediocre (unimodal) foundation models work?!

Bild
Anshul Kundaje@anshulkundaje.bsky.social · last yr.

Awesome paper. Simple post-hoc trick (averaging over homologs) with elegant evolution theory dramatically improves zero shot coding variant effect prediction in pLMs & actually delivers better results from the larger models (inverting the trend with raw likelihoods). 1/

bioRxivpreprint@biorxivpreprint.bsky.social · last yr.

From Likelihood to Fitness: Improving Variant Effect Prediction in Protein and Genome Language Models https://www.biorxiv.org/content/10.1101/2025.05.20.655154v1

Anshul Kundaje@anshulkundaje.bsky.social · last yr.

My sympathy and support for all the international students and postdocs at Harvard. I can only imagine how scary this must feel. Stay strong. And if there's anything I can do to help those in compbio, please reach out. All of us need to rally to help them out. 1/

Peter Koo@pkoo562.bsky.social · last yr.

The premier conference on Machine Learning for Computational Biology is Sep 9-10 at the NY Genome Center in NYC! Submission deadline is June 1 for 2-page abstracts and 8-page papers (eligible for proceedings track). Registration is now open! (Link below) Please retweet!

Bild