TSUMUGI: a platform for phenotype-driven gene network identification from comprehensive knockout mouse phenotyping data

SummaryDeciphering complex organismal phenotypes requires elucidation of coordinated functions among multiple genes, yet this remains a fundamental challenge in functional genetics. The International Mouse Phenotyping Consortium (IMPC) has recently established a comprehensive phenotypic atlas based on systematic single-gene knockout mouse lines, providing an unprecedented resource for gene-phenotype associations in mammals. However, extracting phenotype-associated multiple gene relationships remains challenging. Here, we present TSUMUGI, a platform that identifies a phenotype-driven gene network by leveraging gene-phenotype associations from the IMPC. TSUMUGI enables users to initiate analyses from a user-specified phenotype or gene list and interactively explore a gene network. In addition, TSUMUGI highlights human disease-associated genes, supports flexible network filtering, and allows seamless export of a gene network for downstream analyses. By linking shared phenotypic signatures to putative functional gene modules, TSUMUGI provides a framework for systematic interpretation and hypothesis generation of complex organismal phenotypes. Availability and implementationThe web application is available online at https://larc-tsukuba.github.io/tsumugi/. The command-line tools are distributed via PyPI (https://pypi.org/project/TSUMUGI) and Bioconda (https://bioconda.github.io/recipes/tsumugi/README.html). Source code and documentation are hosted at GitHub (https://github.com/akikuno/TSUMUGI-dev) and archived on Zenodo (https://doi.org/10.5281/zenodo.18464478) under the MIT license.