Back

Discovering cell types underlying rare disease phenotypes using scRNA-seq data from non-diseased tissues

Novoa, J.; Pazos, F.; Chagoyen, M.

2025-12-11 systems biology
10.64898/2025.12.09.693155 bioRxiv
Show abstract

Despite their low individual prevalence, rare diseases collectively pose a significant health burden, affecting millions of people worldwide. These conditions often result from single-gene mutations, yet the cellular contexts in which these alterations act remain largely unknown--information crucial for improving diagnosis and treatment. As patient-derived samples are scarce, we use single-cell RNA sequencing (scRNA-seq) data from non-diseased tissues to identify relevant cell populations. We introduce Cell4Rare, a computational framework that integrates these healthy scRNA-seq datasets with known phenotype-associated genes to map disease phenotypes to specific cell types. Applied across diverse tissues and phenotypes, Cell4Rare was validated against literature-based associations, achieving robust performance with an AUC of 0.71. These results highlight the potential of computational analyses of non-diseased scRNA-seq data to uncover the cellular basis of rare disease phenotypes, paving the way for improved diagnostics and therapeutic strategies.

Published in Briefings in Bioinformatics (predicted rank #27) · training set

Matching journals

The top 7 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.