VarSome is a search engine, aggregator, and impact analysis tool for human genetic variation, designed to help researchers and clinicians find information on genetic variants. It allows users to search for variants by gene name, transcript symbol, genomic location, variant ID, or HGVS nomenclature. VarSome collects data from 30 external databases, including ClinVar, gnomAD, and UniProt, and stores over 33 billion data points describing more than 500 million variants. The system uses MolecularDB, an efficient data warehouse adapted to genomics, and thalia, a tool written in C++ that maps variants to genomic locations and identifies equivalent variants. The front end is built with HTML5, JavaScript (React), and the back end with Python 3 (Django) and C++. VarSome provides a custom genome browser that displays genomic context, including exonic structure, functional domains, and nearby structural variants. It also includes a variant pathogenicity classifier based on ACMG guidelines and integrates data from multiple sources, including population frequency data from gnomAD, pathogenicity predictions from dbNSFP and DANN, and clinically relevant information from CGD. Users can submit their own annotations, linking variants to phenotypes, diseases, or articles, and these annotations are linked to the user's profile. VarSome also provides permanent web links to each variant, allowing users to share findings or refer to variants in publications. Since its release in May 2016, VarSome has grown to 56,000 users from over 120 countries and has been integrated into other websites, including Variant Validator. It serves as a powerful tool for annotating and searching human genomic variants and a platform for sharing knowledge on specific variants.VarSome is a search engine, aggregator, and impact analysis tool for human genetic variation, designed to help researchers and clinicians find information on genetic variants. It allows users to search for variants by gene name, transcript symbol, genomic location, variant ID, or HGVS nomenclature. VarSome collects data from 30 external databases, including ClinVar, gnomAD, and UniProt, and stores over 33 billion data points describing more than 500 million variants. The system uses MolecularDB, an efficient data warehouse adapted to genomics, and thalia, a tool written in C++ that maps variants to genomic locations and identifies equivalent variants. The front end is built with HTML5, JavaScript (React), and the back end with Python 3 (Django) and C++. VarSome provides a custom genome browser that displays genomic context, including exonic structure, functional domains, and nearby structural variants. It also includes a variant pathogenicity classifier based on ACMG guidelines and integrates data from multiple sources, including population frequency data from gnomAD, pathogenicity predictions from dbNSFP and DANN, and clinically relevant information from CGD. Users can submit their own annotations, linking variants to phenotypes, diseases, or articles, and these annotations are linked to the user's profile. VarSome also provides permanent web links to each variant, allowing users to share findings or refer to variants in publications. Since its release in May 2016, VarSome has grown to 56,000 users from over 120 countries and has been integrated into other websites, including Variant Validator. It serves as a powerful tool for annotating and searching human genomic variants and a platform for sharing knowledge on specific variants.