code

A curated set of open-source projects associated with my research on self-supervised learning, multimodal foundation models, and generative dynamics for single-cell omics.

These are the open-source repositories that accompany my publications. Each card links to the GitHub project. For the full list, see my GitHub profile.

microsoft
cell_synergy

Synergistic Information Score (SIS) and benchmark of fusion methods for multimodal cell foundation models.

Beyond alignment, bioRxiv 2026 Python PyTorch Multimodal
theislab
CellHermes

Biological language model that unifies multimodal omics tasks into natural-language question-answer pairs.

CellHermes, bioRxiv 2025 Python LLM LoRA
theislab
CFGen

Flow-matching generative model for multi-modal, multi-attribute single-cell counts.

ICLR 2025 Python Flow Matching scRNA-seq
theislab
ssl_in_scg

Systematic study of self-supervised learning in single-cell genomics; masked-prediction transfers best.

Nature Machine Intelligence 2024 Python PyTorch SSL
theislab
pdac_analysis

Single-cell analysis of PDAC branched-organoid models, heterogeneity, and treatment response.

Nature Biomedical Engineering 2024 Python scRNA-seq PDAC