Network-based computational pipeline for studying variability of transcriptome profiles for human diseases

Abstract

Machine learning applications to high-throughput data in medicine– one of the biggest resources for understanding complex diseases– have been limited thus far. Here, we present a computational approach for assessing the intrinsic variability in the most prominent data type, transcriptomics data for diseasecohorts. Our study looks at situations where multiple data sets for the same disease are available. We leverage concepts of network medicine to assess how the match between a biological network and a set of differentially expressed genes varies across different networks and experiments. Our results showed that different biological networks yielded markedly different results; also, the clustering of diseases depended strongly on the choice of the parameters that were contained in the data analysis and network processing.