Back

Sample-specific network analysis identifies gene coexpression patterns of immunotherapy response in advanced kidney cancer

Yin, L.; Traversa, P.; Elati, M.; Moreno, Y.; Marek-Trzonkowska, N.; Battail, C.

2024-11-04 bioinformatics
10.1101/2024.11.02.621068 bioRxiv
Show abstract

Immunotherapies have recently emerged as a standard of care for advanced cancers, offering remarkable improvements in patient prognosis. However, only a small proportion of patients respond to the treatment, and no definitive molecular hallmark has been identified for clinical use in predicting treatment outcomes. Here, we propose a sample-specific weighted gene network approach to investigate the heterogeneity of patient clinical outcomes by leveraging multiscale network features derived from gene expression data. Our results show that patients exhibiting similar clinical benefits share comparable gene coexpression patterns. Increased gene connectivity and stronger negative gene-gene associations are also pivotal factors in patients with poor prognoses. Moreover, integrating pathway genes with topological measures enables the identification of the perturbed regulation of biological pathways associated with treatment responses. Additionally, sample-level network features enhance the prediction performance of gene expression values-based machine learning models. Collectively, our approach provides valuable guidance on the use of gene network information to stratify cancer patients and to optimize treatment strategies.

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.