XYomics: detecting sex-dependent molecular mechanisms in omics data.

Understanding sex-dependent differences in disease risk, manifestation, and treatment response is essential for precision medicine. While funding agencies now mandate consideration of Sex as a Biological Variable (SABV), existing bioinformatics tools lack systematic approaches to characterize sex-related molecular mechanisms. Current practices frequently treat sex as a confounding variable, which may obscure important biological differences such as sex-specific alterations, sex-dimorphic changes (opposite effects between sexes), and sex-modulated changes (different effect magnitudes). We present XYomics, an open-source R package for systematic analysis of sex-dependent alterations in biomedical omics data. The software identifies sex-specific, sex-dimorphic, and sex-modulated changes at both individual feature and systems levels. XYomics implements dual analytical modes: sex-disease interaction term modeling for adequately powered datasets and sex-stratified analysis with robust non-significance filtering for smaller sample sizes. Using single-cell RNA sequencing data from Alzheimer's disease patients, we demonstrate how XYomics identifies sex-dimorphic genes largely undetected by standard sex-averaged analyses. By integrating statistical categorization with pathway enrichment and network analysis using a curated hormone signaling interactome, the software facilitates discovery of sex-specific biomarkers and disease mechanisms frequently obscured in sex-aggregated analyses.