Understanding the functional consequences of genomic variants relies on multiple lines of evidences. These include among the others the allele frequency of the variant in a healthy population, the predicted effect on the protein, the presence in knowledge bases such as CIViC or ClinVar. While using these criteria some variants can be classified either as benign or disease-causing, in many cases available evidences at the time of analysis are not sufficient to support the classification of the variant. And the number of variants of uncertain significance (VUS) is increasing due to the advances in NGS technology that increase the yield of genetic testing. In practice the higher resolution of NGS testing has produced a greater burden of uncertain results. Variant reinterpretation based on the availability of updated annotations is part of the routine work of research laboratories. In fact, more data are collected about a specific variant, higher the probability to reinterpret the unclassified variant. This task is time consuming as requires the access to multiple sources of information (e.g. software, knowledge bases, literature), which are constantly updated.

To support the interpretation of genetic variants, we developed VariantAlert, a web-based tool to help researchers and clinicians to keep informed about changes in variant annotations extracted from multiple sources. A user can submit one or more lists of variants which are daily re-annotated using external resources accessed through API, such as MyVariant.info annotation API providing links to variant annotations from gnomad, COSMIC, ClinVar, CIViC, and many others.

If a change is detected for the annotation of a variant due to the upgrade of the underlying resource (e.g. change in gnomad allele frequency, presence in COSMIC database, change in ClinVar classification) the user is notified by email and updated annotations are stored on the web-site. Conclusions We expect that VariantAlert will contribute to the interpretation of genetic variants and their classification by keeping the researchers constantly informed of the current content of multiple annotation databases.