MSigDB 2023.1.Mm - GSEA-MSigDB Documentation

MSigDB v2022.1.Mm (Mar 2023)

Important Notices#

This page describes updates made to the Molecular Signatures Database Mouse Collections for release 2023.1 (MSigDB 2023.1.Mm). In order to access the MSigBD mouse collections through the GSEA UI, the GSEA 4.3.0 or newer is required. MSigDB v2023.1 is based on gene annotation data from Ensembl Release 109 (Feb 2023).

Updates to Mouse Collections (MSigDB v2023.1.Mm)#

M1: positional gene sets#

As previously noted in the MSigDB v2022.1.Mm Release Notes the underlying data for the M1 collection remains based on the cytogenetic band annotations provided in the Ensembl 102 release, corresponding to the GRCm38 assembly as cytogentic band annotations for GRCm39 remain unavailable, however gene identifiers have been updated.

M2:CGP#

3 Gene sets contributed by MSigDB users have been added to M2:CGP

M2:CP:Reactome#

Reactome gene sets have been updated to reflect the state of the Reactome pathway architecture as of Reactome v83 (+2 gene sets).
As previously described in the Reactome release notes for MSigDB 7.0, in order to limit redundancy between gene sets within the Reactome sub-collection we applied a filtering procedure based on Jaccard coefficients and distance from the top level of the Reactome event hierarchy.

M2:CP:WikiPathways#

WikiPathways gene sets have been updated to the February 10, 2023 release (+2 gene sets).

M3:GTRD#

GTRD data was updated to the 21.12 release (+7 gene sets).

M5:GO (Gene Ontology)#

Gene sets in these sub-collections are derived from the controlled vocabulary of the Gene Ontology (GO) project: The Gene Ontology Consortium. Gene Ontology: tool for the unification of biology (Nature Genet 2000). The gene sets are named by GO term and contain genes annotated by that term. This collection has been updated to the most recent GO annotations as present in the GO-basic obo file released on 2023-01-01 and NCBI gene2go annotations downloaded on 2023-02-10.

This collection is divided into three sub-collections:

BP: GO Biological process (+67 gene sets). Gene sets derived from the Biological Process Ontology, which are prefixed with "GOBP_".
CC: GO Cellular component (-11 gene sets). Gene sets derived from the Cellular Component Ontology, which are prefixed with "GOCC_".
MF: GO Molecular function (+57 gene sets). Gene sets derived from the Molecular Function Ontology, which are prefixed with "GOMF_"..

These updates were generated in accordance with the procedure described in the GO release notes for MSigDB 7.0.

M8 cell type signature gene sets#

Added gene sets describing uterine cell type identity signatures from Zhang, et al. 2022 Digital Cell Atlas of Mouse Uterus: From Regenerative Stage to Maturational Stage. (+18 gene sets)

CHIP file updates#

MSigDB 2023.1.Mm gene annotations and gene mapping CHIP files have been updated to data from Ensembl 109.
Gene orthology annotations for mapping human and rat genes to their best match mouse orthologs have been updated to Alliance of Genome Resources orthology database release 5.3.0 (2022-10-28)

SQLite Database#

With this release we have created a new SQLite database for the fully annotated gene sets in both the Human (2023.1.Hs) and the Mouse (2023.1.Ms) resources. Each ships as a single-file database usable with any compliant SQLite client. This new format brings the MSigDB contents and metadata with all of the searchability and manipulative power of a full relational database. See our documentation for more details on the contents and usage.

Known Issue: Incidental extra Gene Symbol and Source Member records#

When building MSigDB we create a somewhat larger database and then remove a number of gene sets based on various criteria. The most important of those is a size-based threshold, where gene sets with fewer than 5 or more than 2000 members mapped to gene symbols are removed, but there are others as well.

Well after the release, it was discovered (in July 2023) that certain tables related to Gene Symbols and Source Members had not been properly purged after removal of such gene sets, resulting in the retention of unrelated records. We have created a patched database file that removes these extra records (available on our Download page).

SQLite can be somewhat loose when it comes to referential integrity and allows situations like this to exist even when foreign keys are enabled. This won't cause a problem in most cases where queries are based on the gene_set table, but for those users with the original database it's still a good idea to remove these extra records. This can be accomplished with the following SQL commands from the official SQLite command line shell:

    DELETE FROM gene_set_details WHERE gene_set_id NOT IN (SELECT id FROM gene_set);
    DELETE FROM gene_set_source_member WHERE gene_set_id NOT IN (SELECT id FROM gene_set);
    DELETE FROM source_member WHERE id NOT IN (SELECT distinct(source_member_id) FROM gene_set_source_member);
    DELETE FROM gene_set_gene_symbol WHERE gene_set_id NOT IN (SELECT id FROM gene_set);
    DELETE FROM gene_symbol WHERE id NOT IN (SELECT gene_symbol_id FROM gene_set_gene_symbol);
    VACUUM;
    PRAGMA optimize;

That final DELETE has no actual effect on the database since it turns out that all gene symbol records are referenced by some remaining gene set. It's included here anyway since that might not always be the case in the future, should this issue ever come up again.

It is recommended to do this without turning on the Foreign Key PRAGMA as it will otherwise take an immensely long time to complete. With the FK enforcement off it runs very quickly.

The last two commands are optional but will optimize the size and access patterns. It's probably a marginal benefit but it's worth doing as a one-time operation.

We have decided not to update the files on the server to avoid confusion about contents compared to those who might have already downloaded the files. We might find another way to avoid this in the future, in which case these Release Notes will be updated.