Single-cell high-throughput sequencing, a major breakthrough in life sciences, allows us to access the integrated molecular profiles of thousands of cells in a single experiment. This abundance of data provides tremendous power to unveil unknown cellular mechanisms. However, single-cell data are so massive and complex that it has become challenging to give clues to their underlying biological processes.
The machine learning for integrative genomics G5 group works at the interface of machine learning and genomics, developing methods exploiting the full richness and complementarity of the available single-cell data to derive actionable biological knowledge.
 |
 |
 |
 |
|---|---|---|---|
| MOWGLI: Integrating paired multimodal single-cell data | scConfluence: Integrating unpaired multimodal single-cell data | HuMMuS: Molecular mechanisms from multi-omics single-cell data | ReCoN: Exploring multicellular coordination from single-cell gene expression / multi-omics using mutlilayer network representations |
    |
|
 |
 |
 |
 |
 |
 |
| stories: Learning cell fate landscapes from spatial transcriptomics using Fused Gromov-Wasserstein | scPrint: Transcriptomic foundation model for gene network inference and more | scPRINT-2: Towards the next-generation of cell foundation models and benchmarks | MOMIX: Benchmark of multi-omics joint Dimensionality Reduction (jDR) approaches in cancer study |
| |
|
|
|
 |
 |
 |
 |
|---|---|---|---|
| scDataLoader: a dataloader to work with large single cell datasets from lamindb | benGRN: Awesome Benchmark of Gene Regulatory Networks | GRnnData: Awesome GRN enhanced AnnData toolkit | CIRCE: Predict cis-regulatory interactions between DNA regions |
| codebase · PyPI |
codebase · PyPI |
codebase · PyPI |
codebase · PyPI · manuscript |