Results

VarImpact is a tool to annotate mutations in texts that are mentioned together with a specific impact, pertaining to a change in protein function, cellular localization, exnzymatic activity, active sites, and so on, driven by the mutation. Those impacts can affect behavior on gene, mRNA, protein, cell, or organism level.

Dfferent categories of impacts include:
- gene expression — the variant has an effect on the expression of the gene;
- target — the variant changes a target site (e.g., phosphorylation or RNAi target site);
- binding — the variant affects the binding of the protein to another protein or DNA;
- disease — the variant was observed in patients with a certain disease;
- population — the variant was observed (and conclusions drawn) in a given population (could be ethnicity);
- sensitivity — the variant changes sensitivity of the gene product to a drug, temperature, or other stress such as radiosensitivity;
- substance concentration — the variant affects how this gene product or another downstream substance (especially for mutated enzymes) accumulates;
- cellular location — the variant affects the subcellular localization of the protein;
- protein stability — the variant has an effect on the stability of the protein or the mRNA;
- biomarker — the variant was used as a molecular marker for a certain indication;
- protein function — the variant affects the protein's function (other than any of the functions above);
- downstream activation — the variant affects the activation of a downstream gene, relevant mostly in conjunction transcription factors;
- driver mutation — the variant was reported to be driving a given indication or as a risk factor;
- cell — the variant affects cell invasiveness, cell adhesion, apoptosis, etc.

VarImpact is meant to crawl extensive libraries, such as PubMed abstracts or full text scientific articles, and extract genes, mutations, and their impacts. All annotations are provided in a database, and can be retrieved from any other platform to annotate genes and genetic variants with observed and reported "phenotypic" impacts. The data thus serve as additional information to enrich functional categorizations or predictions, such as provided by SIFT, Polyphen2, and similar tools (see dbNSFP for more); mutations frequencies observed in diseases populations versus normal; and so on.

VarImpact also comes with a simple webinterface to search the database by gene, impact category, keywords, and a few other criteria. The web interface requires an Apache Webserver and an RDBMS such as MySQL or Oracle.