Back

Decomposing patient heterogeneity of single-cell cancer data by cross-attention neural networks

Subedi, S.; Park, Y. P.

2025-06-06 oncology
10.1101/2025.06.04.25328900 medRxiv
Show abstract

Gene expression variation in cancer cells is attributed to many inherited and environmental factors, including genetic variants and cellular landscapes. Decomposing different sources of information is intractable with single-cell RNA-seq alone. However, we show that our new approach can split them with the help of multiple patients, assuming that cell types are widely shared and genetic effects are specifically present in a particular patient. Our approach based on a cross-attention neural network was applied to three different cancer types to identify cell types and patient-specific genetic effects in transcriptomic data. Residual expressions, excluding cell types, can implicate patient-specific disease mechanisms.

Matching journals

The top 5 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.