2023 ACMG Annual Clinical Genetics Meeting Digital Edition: Poster Gallery-The Impact of Machine Learning Algorithms in Reducing VUS for Individuals from Underrepresented Populations Compared to Well Studied Populations

The Impact of Machine Learning Algorithms in Reducing VUS for Individuals from Underrepresented Populations Compared to Well Studied Populations

Back to course

Pdf Summary

This document discusses the impact of machine learning algorithms on reducing variants of uncertain significance (VUS) for individuals from underrepresented populations compared to well-studied populations. The document includes several references and disclosures from Invitae, a company specializing in genetic testing.<br /><br />The document mentions various models and types of evidence used in the study. These include physicochemical conservation, homolog information, evolutionary models of variant effect (EVE), multiplex assays of variant effects (MAVEs), SpliceAI splicing prediction, cellular evidence modeling, gene-specific engine, molecular stability engine, multidimensional hotspots, and population frequency modeling using gnomAD.<br /><br />Figure 2 shows that out of approximately 159,000 individuals, 45% had machine learning evidence applied to at least one variant. Machine learning evidence contributed to the classification of pathogenic or likely pathogenic variants in 24% of individuals for both benign/likely benign and pathogenic/likely pathogenic variants. Less than 1% of individuals had machine learning evidence contributing to the classification of pathogenic/likely pathogenic variants alone.<br /><br />Overall, the document highlights the use of machine learning algorithms in genetic testing and their potential to reduce VUS for individuals from underrepresented populations.

Asset Subtitle

Presenting Author - Britt A. Johnson, PhD, FACMG; Co-Author - Ana Morales, MS, CGC; Co-Author - Flavia M. Facio, MS, CGC; Co-Author - Laure Fresard, PhD; Co-Author - Dianalee McKnight, PhD, FACMG; Co-Author - Yuya Kobayashi, PhD; Co-Author - Jason Reuter, PhD; Co-Author - Keith Nykamp, PhD; Co-Author - Swaroop Aradhya; Co-Author - Alexandre Colavin, PhD;

Meta Tag

Genetic Testing

NextGen Sequencing

Sequencing

Co-Author Ana Morales, MS, CGC

Co-Author Flavia M. Facio, MS, CGC

Co-Author Laure Fresard, PhD

Co-Author Dianalee McKnight, PhD, FACMG

Co-Author Yuya Kobayashi, PhD

Co-Author Jason Reuter, PhD

Co-Author Keith Nykamp, PhD

Co-Author Swaroop Aradhya

Co-Author Alexandre Colavin, PhD

Presenting Author Britt A. Johnson, PhD, FACMG

Keywords

machine learning algorithms

variants of uncertain significance

underrepresented populations

well-studied populations

physicochemical conservation

homolog information

evolutionary models of variant effect

SpliceAI splicing prediction

cellular evidence modeling

genetic testing