Back

Multimodal weakly supervised learning to identify disease-specific changes in single-cell atlases

Litinetskaya, A.; Shulman, M.; Hediyeh-zadeh, S.; Moinfar, A. A.; Curion, F.; Szalata, A.; Omidi, A.; Lotfollahi, M.; Theis, F. J.

2024-07-29 bioinformatics Community evaluation
10.1101/2024.07.29.605625 bioRxiv
Show abstract

To deliver clinically relevant insights from large patient cohorts profiled with single-cell technologies, a key challenge is to relate sample-level and single-cell measurements. We present MultiMIL, a deep learning framework that applies attention-based multiple-instance learning for phenotype prediction and cell state identification. We applied MultiMIL to peripheral blood mononuclear cells from COVID-19 patients, the Human Lung Cell Atlas, and a spatial proteomics breast cancer dataset, demonstrating how our model can be utilized to find phenotype-associated cell states, learn phenotype-informed sample representations, and expand disease signatures.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.