This app includes Human MSigDB 2026.1.Hs Hallmark, complete C2 canonical pathways (C2:CP), and Gene Ontology (GO:BP, GO:CC and GO:MF). Gene sets are repackaged from the official Gene Symbol GMT files without changing membership. Unmatched genes are omitted from scoring and reported to the user.
MSigDB copyright © 2004–2026 Broad Institute, Inc., Massachusetts Institute of Technology, and Regents of the University of California. MSigDB data is provided generally under Creative Commons Attribution 4.0 International; see the official license terms.
KEGG MEDICUS gene sets (KEGG_MEDICUS_ prefix): copyright © 2009–2023 Kanehisa Laboratories, all rights reserved; provided under CC BY-SA 4.0.
KEGG LEGACY gene sets (KEGG_ prefix without MEDICUS): copyright © 1995–2017 Kanehisa Laboratories, all rights reserved. Included in MSigDB under license to the Broad Institute with qualified permission to include in this release. The copyright holder's academic and commercial licensing policies and KEGG copyright policies apply; inclusion requires retention of this copyright acknowledgment and compliance with those policies.
BioCarta gene sets (BIOCARTA_ prefix): copyright © 2000–2017 BioCarta, all rights reserved. Included in MSigDB under license to the Broad Institute with qualified permission to include in this release, subject to BioCarta's Disclaimer of Liability and of Warranties. Inclusion is conditioned upon retention of this copyright acknowledgment and agreement to that disclaimer. These additional terms are retained from the official MSigDB terms.
Original source: MSigDB 2026.1.Hs release. Individual gene-set links are retained in the packaged catalog.
References: Subramanian et al. (2005), PNAS, doi:10.1073/pnas.0506580102; Liberzon et al. (2011), Bioinformatics, doi:10.1093/bioinformatics/btr260; Liberzon et al. (2015), Cell Systems, doi:10.1016/j.cels.2015.12.004; Castanza et al. (2023), Nature Methods, doi:10.1038/s41592-023-02014-7.
Module scoring follows Seurat AddModuleScore (Seurat 5.3.0); method reference: Tirosh et al. (2016), Science, doi:10.1126/science.aad0501. The app uses a block-based implementation to keep expression data on disk.