Artificial intelligence (AI) is currently a buzzword across almost all scientific disciplines and has the potential to revolutionize diagnostic approaches in inherited diseases. But when it comes to variant curation, is AI capable of replacing human expertise?

A new study by Stanford University compares data quality from their Automatic VAriant evidence DAtabase (AVADA) to the Human Gene Mutation Database (HGMD). The results show that expert curation by variant scientists is irreplaceable in the field of clinical genomics, with HGMD recovering nearly 40% more disease-associated mutations than the AI-powered AVADA.

In this webinar, we demonstrate the depth and breadth of HGMD through a series of use-cases where the manually curated database outperforms AI solutions, including how HGMD:

• Provides genomic coordinates and Human Genome Variation Society (HGVS) nomenclature for variants as well as citations and comments for key publications
• Indicates where a variant is described in a paper or supplemental text
• Enables database entries to be visualized via tools, such as IGV and UCSC Genome Browser
• Simplifies literature review with legacy/alternative variant naming schemes
• Supports CNV interpretation and “reverse” look-up by phenotype