PHARMACOGENE STAR ALLELE CALLING FROM WHOLE GENOME SEQUENCING POPULATION DATASETS: IDENTIFICATION OF AN EFFICIENT TOOL

Abstract

Pharmacogenomics (PGx) plays a crucial role in the precision and personalized medicine by identifying the genetic variations that affect drug response. One common approach to determine the genotype of pharmacogenes and predict metabolizer phenotype status is star allele calling. Star alleles are the combination of genetic variants in a gene that contributes to a functional effect. In this study, we conducted a comparative analysis of different bioinformatics tools commonly used for star allele calling in PGx research. Our analysis included tools such as PharmCAT, Aldy, and Cyrius which are widely used for population PGx analysis by different biobanks. We used each tool to determine the genotype/diplotype of pharmacogenes associated with drug metabolism pathways in the Qatari population genomic data. We compared the results from each tool and looked for the concordance between them. Additionally, we evaluated the performance of each tool with respect to Genetic Testing Reference Material (GeT-RM) by calculating the Positive Predictive Value (PPV). PharmCAT, Aldy and Cyrius showed 97%, 93% and 97% precision respectively while genotyping GeT-RM samples. Our findings revealed that there is 77.2% concordance between Aldy and PharmCAT while calling star alleles for pharmacogenes in the large population dataset. We identified that PharmCAT and Cyrius use less computational resources, processing time and generate clinically interpretable reports whereas Aldy is user friendly and good for large population dataset. Overall, this comparative analysis provides valuable insights into the strengths and limitations of different bioinformatics tools used for star allele calling in PGx.

Date of Award	2024
Original language	American English
Awarding Institution	HBKU College of Health & Life Sciences