SIDISH integrates single-cell and bulk transcriptomics to identify high-risk cells and guide precision therapeutics through in silico perturbation.

Abstract

Single-cell RNA sequencing (scRNA-seq) provides high-resolution insights into cellular heterogeneity but remains costly, restricting its use to small cohorts that often lack comprehensive clinical data, reducing translational relevance. In contrast, bulk RNA sequencing is scalable and cost-effective but obscures critical single-cell insights. We introduce SIDISH, a neural network framework that integrates the granularity of scRNA-seq with the scalability of bulk RNA-seq. Using a variational autoencoder, deep Cox regression, and transfer learning, SIDISH identifies high-risk cell populations while enabling robust clinical predictions from large-cohort data. Its in silico perturbation module identifies therapeutic targets by simulating interventions that reduce high-risk cells associated with adverse outcomes. SIDISH also generalizes to spatial transcriptomics, identifying high-risk cells and mapping them within their native tissue microenvironment. Applied across diverse diseases, SIDISH establishes the link between cellular dynamics and clinical phenotypes, facilitating biomarker discovery and precision medicine. By unifying single-cell insights with large-scale clinical data, SIDISH advances computational tools for disease risk assessment and therapeutic prioritization, offering an integrative and scalable approach to precision medicine.