Bioinformatics Seminar
Time: 11AM
Venue: Davis Auditorium and Teams
27 September 2022
Improving Clinical Translation of High-throughput Functional Assay Data with MaveDB
Alan RubinWEHI Bioinformatics
Understanding the effects of genetic variants is essential for using a patient’s sequence to guide diagnosis and treatment. However, our ability to acquire sequence data vastly outstrips our ability to interpret it, leading to the rapid accumulation of Variants of Uncertain Significance (VUS) in diagnostic datasets. High-quality functional assay data is important evidence for interpreting VUS, but traditional functional assays only evaluate a handful of variants at a time. By contrast, Multiplexed Assays of Variant Effect (MAVEs) deliver accurate, ancestry-agnostic functional data for thousands of variants in a gene simultaneously, making them a powerful tool for variant reclassification, and data generation continues to accelerate.This talk will describe MaveDB, the database of record for MAVE datasets, and ongoing efforts to enable the clinical translation of MAVE data. This work includes the definition of purpose-built data models and standards to enable integration into clinical workflows that will reduce the number of VUS returned clinically in Australia and internationally. MaveDB already contains more than 250 datasets and 3 million variant effect measurements across diverse targets, including clinically actionable genes such as BRCA1, MSH2, PTEN, SCN5A and TP53. The field continues to grow rapidly, both in the number of genes assayed and the scale of these studies, presenting new opportunities for clinicians and researchers alike, particularly in the application of machine learning and advanced statistical methods.