PAnno: A pharmacogenomics annotation tool for clinical genomic testing

Area of Science:

• Genomic Medicine
• Pharmacogenomics
• Clinical Decision Support

Background:

• Next-generation sequencing (NGS) is crucial for clinical genomic testing of drug response phenotypes.
• Short reads in NGS present challenges for accurate diplotype inference, potentially limiting genotype-guided drug therapy.
• Accurate diplotype determination is essential for personalized medicine and optimizing drug efficacy and safety.

Purpose of the Study:

• To develop and validate an automated tool, PAnno, for accurate diplotype inference and pharmacogenomic annotation.
• To enhance the effectiveness of genotype-guided drug therapy by overcoming NGS limitations.
• To provide an end-to-end solution for clinical pharmacogenomics decision support.

Main Methods:

• Implemented an automated Pharmacogenomics Annotation tool (PAnno) to parse germline variant call format (VCF) files.
• Developed a ranking model for diplotype inference based on allele definition and population allele frequency.
• Validated PAnno's performance against existing tools using Genetic Testing Reference Materials Coordination Program (GeT-RM) data.
• Integrated Clinical Pharmacogenetics Implementation Consortium (CPIC) and Pharmacogenomics Knowledgebase (PharmGKB) annotations for prescribing recommendations and phenotype prediction.

Main Results:

• PAnno accurately infers diplotypes by incorporating allele definitions and population frequencies.
• The tool provides clear prescribing recommendations, classifying drugs into "avoid use," "use with caution," and "routine use" categories.
• PAnno predicts drug phenotypes related to toxicity, dosage, efficacy, and metabolism.
• Performance validation demonstrated PAnno's predictive accuracy compared to other tools.

Conclusions:

• PAnno offers a robust, automated solution for clinical pharmacogenomics.
• The tool addresses the challenge of diplotype inference from NGS data, improving genotype-guided therapy.
• PAnno facilitates informed clinical decisions by resolving, annotating, and reporting germline variants.