Multiomics
Reading Group
Exploring the future of ML modeling for multi-omics data.
(Mostly) every Wednesday, 11 AM–12 PM EST.
Where Biology Meets Compute
The overlooked signal in the biology. The causal framework that shouldn't work but does. The training trick, the generative model, the insight that reshapes what the next architecture needs to learn. Both sides of the bench welcome.
Genomics & Transcriptomics
Foundation models for DNA/RNA, language models for sequences, single-cell expression modeling.
Proteomics & Structure
Protein language models, structure prediction, intermolecular interactions, and multi-modal bio representations.
Model Design & Training
Tokenization strategies, pre-training/fine-tuning regimes, architecture choices that actually matter.
Multi-Omics Integration
Bridging single- to multi-omics. Cross-modal learning. Making different data types talk to each other.
Evaluation & Benchmarks
How do we know any of this works? Perturbation prediction, benchmark design, and honest assessment.
Position Papers & "Why"
The hard questions. Why build foundation models for biology? What should they even do? We read the arguments.