RNA-seq Titration Results supporting "Cross-platform normalization enables machine learning model training on microarray and RNA-seq data simultaneously"

This data accompanies the manuscript "Cross-platform normalization enables machine learning model training on microarray and RNA-seq data simultaneously" by Foltz, Taroni, and Greene. Please refer to our github page.

The file contains all results and outputs from the analysis pipeline as well as intermediate files (including models and normalized data) from one repeat (seed 3274).

Abstract: Large compendia of gene expression data have proven valuable for the discovery of novel biological relationships. Historically, the majority of available RNA assays were run on microarray, while RNA-seq has become the platform of choice for many new experiments. The data structure and distributions between the platforms differ, making it challenging to combine them. Here we perform supervised and unsupervised machine learning evaluations, as well as differential expression analyses, to assess which normalization methods are best suited for combining microarray and RNA-seq data. We find that quantile and Training Distribution Matching normalization allow for supervised and unsupervised model training on microarray and RNA-seq data simultaneously. Nonparanormal normalization and z-scores are also appropriate for some applications, including differential expression analysis. We demonstrate that it is possible to perform effective cross-platform normalization and combine microarray and RNA-seq data for machine learning applications.