Discovery and Variant Effect Prediction with the next generation of ProtVar

Genomic variation can perturb normal biological processes through diverse, context-dependent mechanisms, making multidimensional data integration essential for interpreting variant effects. While millions of missense variants exist within the human proteome, only a tiny fraction currently possess clinical annotations.

ProtVar, developed by the Protein Function Development at EMBL-EBI, is a human variant annotation and assessment resource that bridges genomic coordinates, protein sequences, structures, and functional data. It aggregates curated insights and predictions from sources including UniProt, Ensembl, PDBe, Open Targets, and AlphaFoldDB.

Having recently evolved from a single variant lookup tool into a proteome-wide discovery engine, ProtVar now allows researchers to search, filter, and contextualise over 500 million potential human missense variants. Featuring new AI-powered real-time semantic search, ProtVar maps plain language queries such as disease terms, pathways, or drug responses e.g., "Warfarin sensitivity", directly to functional and structural mechanisms.

This webinar will provide practical guidance on using ProtVar to explore missense variation. It will feature live demonstrations of its new discovery features, including real-time AI searching, multi-criteria filtering across all potential variants in the proteome, structural visualisations, and data export options.

This event is part of the broader "UniProt 2026-2027" series. You can follow the link for more information about the series and its webinars.

Who is this course for?

This webinar is suitable for anyone interested in interpreting and discovering the biological effects of human missense variants. More specifically, it can be of interest for:

• Clinical geneticists and clinicians: prioritising patient variants associated with developmental disorders or rare genetic diseases
• Drug discovery researchers and protein scientists: investigating variation in drug binding pockets, active sites, and target protein stability
• Bioinformaticians and computational biologists: Seeking to retrieve or programmatically integrate large scale variant annotations via the ProtVar REST API

No prior computational or bioinformatics experience is required. However, an undergraduate-level knowledge of molecular biology and genetics is recommended.

Outcomes

By the end of this webinar, you will be able to:

• Search ProtVar using natural-language concepts, disease phenotypes and drug terms
• Apply functional and structural filters to prioritise relevant variants
• Interpret the structural and functional effects of variants using integrated prediction tools and visualisations
• Export and integrate ProtVar data using web downloads and the REST API

This webinar, hosted on Zoom Webinar, is free to attend, but you need to register for a place beforehand, using the 'Register' button. Once registered, you will receive a confirmation email with a link to join the webinar.

The webinar will be recorded, and the recording will be made available on this page one day after the event. Please note that we do not issue certificates of attendance for our webinars. However, you will receive an email from our Zoom account confirming your participation a day after the event