🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

How can foundational models + directed evolution make proteins >100x-fold better? We (Omar Abudayyeh, Kaiyi Jiang, Matteo Di Bernardo, Zhaoqing Yan) report in Science Magazine on EVOLVEpro, a general active learning approach to improve protein and enzyme function science.org/doi/10.1126/sc…

Bio foundation models are great design and engg tools. But can they help decode the fundamental principles of life? We harnessed a single-cell FM for decoding the long-debated relationship between genome arch. and gene coregulation. It all started with an idle curiosity. 1/

PISA: a versatile interpretation tool for visualizing cis-regulatory rules in genomic data biorxiv.org/content/10.110…

Meet ATOMICA 🧬 Kudos to stellar Ada Fang Universal AI model that learns atom-scale representations of interactions across all molecular types: small molecules, metal ions, peptides, proteins, DNA, and RNA Stay tuned for our wet lab results, and check out Ada Fang thread 👇

So this interview lasted 2 hours so this “you’re scaring me” part might seem like an overreaction or fearmongering to someone without that context. There’s a lot of evidence to support my hypothesis that a potential H5N1 pandemic would be worse than COVID.

New results for o3 and o4-mini have been added to the ARC Prize leaderboard. Here are some key takeaways: 1/ 🧵

Why do large protein language models like ESM2-15B underperform compared to medium-sized ones like ESM2-650M in predicting mutation effects? 🤔 We dive into this issue in our new preprint—bringing insights into model scaling on mutation effect prediction. 🧬📉

Delighted to share our latest work deciphering the landscape of chromatin accessibility and modeling the DNA sequence syntax rules underlying gene regulation during human development! biorxiv.org/content/10.110…. Read on for more 🧵 [1/16]

Check out this great resource of single cell multiome data, cell annotations, regulatory DNA models of cell type resolved chromatin accessibility & comprehensive annotation of motifs & sequence syntax for multiple tissues in early human development 1/

New publication. We quantify the aggregation of >100,000 random protein sequences to train CANYA, a convolution-attention hybrid neural network to predict aggregation from sequence. With Benedetta Bolognesi Centre for Genomic Regulation (CRG) Wellcome Sanger Institute IBEC science.org/doi/10.1126/sc…

Ever wondered what a snapshot of single cell/spatial dataset can reveal? scCellFie - a scalable tool that goes beyond gene expression, making metabolic analysis possible 3-2-1 Say GENES🤳 BioRxiv doi.org/10.1101/2025.0… TRY NOW sccellfie.readthedocs.io #DataScience #Bioinformatics

We present SELPHI 2.0 a machine learning model integrating >40 sequence, omics and structural features to predict kinase-substrate interactions between 420 kinases and 240K phosphosites and improve interpretation of phosphoproteomics data sciencedirect.com/science/articl…

Very exciting large scale science from Omer Ali Bayraktar Stegle Lab Moritz Mall labs! Comprehensively characterising the variability of GBM cell states and their spatial organisation at single cell multimodal resolution. Part 1 with a focus on spatial organization.

Very exciting large scale science from Omer Ali Bayraktar Stegle Lab Moritz Mall labs! Comprehensively characterising the variability of GBM cell states and their spatial organisation at single cell multimodal resolution. Part 2 with a focus on decoding GBM gene regulation.

What keeps cancer cells in check—and can we tame hem? In 2 collaborative preprints, we trace glioblastoma cell transitions in space and decode what regulates their plasticity 🧠👇 📍 Spatial & multi-omic atlas: biorxiv.org/content/10.110… 📍 Regulatory logic: biorxiv.org/content/10.110…

Designing DNA-binding proteins shouldn’t require structures. ❌🧬 With DPAC, we align protein and DNA language models via contrastive learning -- then use simulated annealing to design new binders straight from sequence! 🔀 📜: biorxiv.org/content/10.110… 💻: github.com/programmablebi…

Nice example of low hanging fruit connecting the dots. Some of the comments with links to papers suggest this is not a denovo discovery & may have been obvious in hindsight. But a lot of things are obvious in hindsight. 1/

Social media is like recurrent intrusive thoughts but generated outside of your brain and algorithmically selected to be more recurrent and more intrusive.

Synthetic biology could enable new types of programmable therapeutics. Our new preprint introduces synthetic protein circuits that selectively trigger cell death in Ras-mutant cancer cells, with interesting advantages compared to existing approaches. biorxiv.org/content/10.110…

I wrote a quick application note on Tomtom-lite, a Python implementation of the Tomtom algorithm for comparing PWMs against each other. This implementation can be 10-1000x faster and, as a Python function, can be integrated into your workflows easier. biorxiv.org/content/10.110…