Bioinformatics Seminar
Time: 11AM
Venue: Davis Auditorium and Online
27 August 2024
A central problem in precision medicine is understanding the effects of individual DNA variants. Measuring their activity in the lab provides valuable evidence for clinical interpretation but can be both time- and resource-intensive. Multiplexed assays of variant effect (MAVEs) are a family of experimental techniques that allow researchers to measure many thousands of variants in a gene or other functional element in parallel, generating a large volume of high-quality functional evidence. To support discovery and enable clinical translation, we developed MaveDB, which has been embraced by the research community as the database of record and become an authoritative source of MAVE data for national and international clinical genomics resources. MaveDB contains 2000 datasets and over 7 million variant effect measurements generated in Australia and across the world. The platform supports the sharing and discovery of bioinformatic reanalysis, imputation, and machine learning-based integration of multiple datasets, and the translation of such results directly into clinical practice. Because genes of high clinical relevance drive MAVE data generation and assay development, there is a clear need for purely computational approaches that can capture generalisable knowledge and apply it to improve our understanding of all genes, the majority of which are lower priority. Due to the size and comprehensive nature of the data in MaveDB, it is a valuable source of training and validation data for AI-based approaches to clinical variant effect prediction, drug target identification, biophysical protein modelling, and other diverse applications.
Alan RubinWEHI Bioinformatics
Enabling clinical translation and modelling of high-throughput functional assay data