SALMON: Survival Analysis Learning With Multi-Omics Neural Networks on Breast Cancer

TLDR

Improved cancer prognosis is a key goal of precision medicine, yet existing models struggle to aggregate and filter predictors from complex data, prompting interest in flexible deep‑learning neural networks that can handle non‑linear relationships. The study aims to identify which data types most improve prognosis by applying deep‑learning networks to assess how gene expression predicts Cox regression survival in breast cancer. SALMON, a deep‑learning framework, aggregates gene expression and biomarker data, reduces raw expression to eigengene modules via co‑expression network analysis, and applies feature selection and enrichment to predict survival and interpret modules. Using more omics data improved model performance, demonstrating the feasibility of discovering breast‑cancer co‑expression modules and providing a blueprint for future deep‑learning survival analyses.

Abstract

Improved cancer prognosis is a central goal for precision health medicine. Though many models can predict differential survival from data, there is a strong need for sophisticated algorithms that can aggregate and filter relevant predictors from increasingly complex data inputs. In turn, these models should provide deeper insight into which types of data are most relevant to improve prognosis. Deep Learning-based neural networks offer a potential solution for both problems because they are highly flexible and account for data complexity in a non-linear fashion. In this study, we implement Deep Learning-based networks to determine how gene expression data predicts Cox regression survival in breast cancer. We accomplish this through an algorithm called SALMON (Survival Analysis Learning with Multi-Omics Neural Networks), which aggregates and simplifies gene expression data and cancer biomarkers to enable prognosis prediction. The results revealed improved performance when more omics data were used in model construction. Rather than use raw gene expression values as model inputs, we innovatively use eigengene modules from the result of gene co-expression network analysis. The corresponding high impact co-expression modules and other omics data are identified by feature selection technique, then examined by conducting enrichment analysis and exploiting biological functions, escalated the interpretation of input feature from gene level to co-expression modules level. Our study shows the feasibility of discovering breast cancer related co-expression modules, sketch a blueprint of future endeavors on Deep Learning-based survival analysis.