Profile avatar
chatgtp.bsky.social
Machine learning for molecular biology. ELLIS PhD student at Fabian Theis lab. EPFL alumnus.
27 posts 1,763 followers 4,094 following
Regular Contributor
Active Commenter
Posts 24 Comments 23

Excited to share that I started my summer at @genentech.bsky.social BRAID Perturbation team in SF with Alex Wu! It's my first time on the West Coast - If you are around and would like to talk about ML and/or biology, hit me up! Looking fwd to the AI x Bio Unconference tomorrow 🚀

submitted 19 hours ago • 0 comments

Analyzing your single-cell data by mapping to a reference atlas? Then how do you know the mapping actually worked, and you’re not analyzing mapping-induced artifacts? We developed mapQC, a mapping evaluation tool www.biorxiv.org/content/10.1... from the ‪@fabiantheis lab. Let’s dive in🧵

submitted 15 days ago • 2 comments

From cell lines to full embryos, drug treatments to genetic perturbations, neuron engineering to virtual organoid screens — odds are there’s something in it for you! Built on flow matching, CellFlow can help guide your next phenotypic screen: biorxiv.org/content/10.1101/2025.04.11.648220v1

submitted 56 days ago • 1 comment

Just a gentle reminder that deceptive hyping in scientific publications (which includes preprints) is actually antithetical to the core mission of the scientific process. We can stay grounded, truthful, humble while being ambitious. Revealing caveats, pitfalls & limitations speeds up progress.

submitted 71 days ago • 3 comments

Our paper benchmarking feature selection for scRNA-seq integration and reference usage is out now www.nature.com/articles/s41...! Keep reading for more about how we did the study and what we found out 🧵 👇 1/16

submitted 92 days ago • 6 comments

Exciting dataset! If you're looking for a complimentary scRNA-seq drug perturbations in healthy/primary tissue (PBMCs), check out our dataset with ~36% # drugs of Tahoe. proceedings.neurips.cc/paper_files/...

submitted 113 days ago • 0 comments

The fact that we seem to be marching straight towards another cold war, where AI is the defining technology, is hard to emotionally accept and even harder to deeply accept how hard some of the next few years can become.

submitted 127 days ago • 1 comment

Last moments of closed-source AI 🪦 : Hugging Face is openly reproducing the pipeline of 🐳 DeepSeek-R1. Open data, open training. open models, open collaboration. 🫵 Let's go! github.com/huggingface/...

submitted 144 days ago • 0 comments

Trying to identify preclinical models that resemble clinical tumors you work on? Check out our MOBER, now out in @science.org Advances! www.science.org/doi/10.1126/... . There's also a web app to explore the results mober.pythonanywhere.com

submitted 147 days ago • 1 comment

When a paper is published, any data must be easily and completely accessible, or the publication is a sham and should be retracted editorially.

submitted 157 days ago • 3 comments

1/7 Planning to build a single-cell atlas? Or wondering how atlases can be useful to your research? Read our guide on single-cell atlases www.nature.com/articles/s41... published in Nature Methods, by @lisasikkema.bsky.social, @khrovatin.bsky.social, Malte Luecken, @fabiantheis.bsky.social et al.

submitted 187 days ago • 1 comment

1/🚀 Excited to share RegVelo, our new cell model combining RNA velocity with gene regulatory network (GRN) dynamics to model cellular changes and predict in silico perturbations. Here's how it works and why it matters! 🧵👇 biorxiv.org/content/10.1101/2024.12.11.627935v1

submitted 188 days ago • 3 comments

What are the most interesting datasets and benchmark-related work for ML in drug discovery at NeurIPS? We’ll be at the conference doing short interviews with researchers and handing out some Polaris merch! Here’s who we have on the shortlist. 🧵

submitted 191 days ago • 2 comments

Public service announcement: Conference deadlines & other artificial deadlines are not excuses to put out half-baked work. If u want to put out half-baked work, u shud clearly state in the paper that results are preliminary & discuss areas of concern that might need careful verification. 1/

submitted 194 days ago • 2 comments

Here's a Cheminformatics Starter pack! Let me know if you would like to be added! And more importantly, do encourage people to join the blue skies :D

submitted 213 days ago • 32 comments

Three BioML starter packs now! Pack 1: go.bsky.app/2VWBcCd Pack 2: go.bsky.app/Bw84Hmc Pack 3: go.bsky.app/NAKYUok DM if you want to be included (or nominate people who should be!)

submitted 197 days ago • 15 comments

submitted 196 days ago • 0 comments

🚨 NeurIPS 2024 Spotlight Did you know we lack standards for AI benchmarks, despite their role in tracking progress, comparing models, and shaping policy? 🤯 Enter BetterBench–our framework with 46 criteria to assess benchmark quality: betterbench.stanford.edu 1/x

submitted 205 days ago • 5 comments

It's been a pleasure to work on the Human Neural Organoid Atlas. Congratulations to the great team across institutes 🎉 ! rdcu.be/d0VBH

submitted 209 days ago • 0 comments

Interested in predicting transcriptomic effects of perturbations? Check out our @NeurIPS24 D&B spotlight living perturbation prediction benchmark & new drug perturbation dataset: - paper: openreview.net/forum?id=WTI... ! - benchmarking platform: openproblems.bio/results/pert... 🧵1/8

submitted 215 days ago • 1 comment

If you're just coming in from Twitter, a great way to reconnect with your community is to find the people you were following there with the Sky Follower Bridge. It works really well! www.sky-follower-bridge.dev

submitted 215 days ago • 2 comments

Thrilled to share our paper on transformers in single cell omics out in @natureportfolio.bsky.social ! Check it out to learn about the latest models, how they are trained, applied, and their limitations. rdcu.be/dQuVW (🧵1/7)

submitted 313 days ago • 1 comment

AlphaMissense looks exciting, but I wish every published paper was reproducible and extensible. "we are not releasing the trained model weights, the code is not meant to be used to reproduce the AlphaMissense scores" www.science.org/doi/10.1126/... github.com/deepmind/alp...

submitted 636 days ago • 1 comment

The University Hospital in Greifswald is looking for an assistant professor in bioinformatics. Greifswald is a beautiful place and we would love to have you here! Please spread the word. #job #bioinformatics www.academics.de/jobs/w1-juni...

submitted 642 days ago • 1 comment