Uploaded October 2025 | Updated September 2026, 3 hours ago
How do you resolve complex pharmacogenes end to end? In this short session, Matt Holt shows how StarPhase uses HiFi long-read whole genome sequencing (WGS) data to enable pharmacogenomic diplotyping.
StarPhase is designed to support diplotyping across a diverse set of pharmacogenes, including variant-defined genes, full-length sequence genes such as HLA, and the structurally complex CYP2D6. Leveraging long-read fragments that are often approximately 15 kb, the workflow helps phase variants into star alleles, assemble full-length HLA haplotypes, and directly observe CYP2D6 copy-number changes. The current release covers 35 genes (23 variant-defined, 11 HLA, plus CYP2D6) (as of August 2025) and produces sample-specific visualizations to support review.
Holt also summarizes concordance with previously reported outputs across about 147 datasets, pedigree checks with no observed inheritance violations in tested families, and population summaries from about 1,450 HiFi WGS datasets across five sites. The talk closes with practical guidance for running StarPhase via a WDL workflow with standard inputs (reference genome, phased VCF, aligned BAM). This video is part of the Spotlight on Human Genomics series focused on practical tools and study-to-insight workflows.
00:00 Introduction and what is PGx diplotyping
02:34 Why long-read sequencing for PGx
05:10 StarPhase overview and gene coverage
06:10 Three algorithms: simple, HLA, CYP2D6
07:31 Visualization: CYP2D6 *4x2/*5 example
09:46 Benchmarking against prior outputs
11:31 Population analyses across ancestries
13:55 Novel alleles and rearrangements
14:55 How to run StarPhase (GitHub, Bioconda)
15:32 Closing
*Subscribe for more PacBio genomics content:*
youtube.com/@PacificBiosciences?sub_confirmation=1
Learn more about PacBio at pacb.com
Legal & Trademarks: Visit pacb.com/legal-and-trademarks
#pharmacogenomics #longreadsequencing #HiFiSequencing
How do you resolve complex pharmacogenes end to end? In this short session, Matt Holt shows how StarPhase uses HiFi long-read whole genome sequencing (WGS) data to enable pharmacogenomic diplotyping.
StarPhase is designed to support diplotyping across a diverse set of pharmacogenes, including variant-defined genes, full-length sequence genes such as HLA, and the structurally complex CYP2D6. Leveraging long-read fragments that are often approximately 15 kb, the workflow helps phase variants into star alleles, assemble full-length HLA haplotypes, and directly observe CYP2D6 copy-number changes. The current release covers 35 genes (23 variant-defined, 11 HLA, plus CYP2D6) (as of August 2025) and produces sample-specific visualizations to support review.
Holt also summarizes concordance with previously reported outputs across about 147 datasets, pedigree checks with no observed inheritance violations in tested families, and population summaries from about 1,450 HiFi WGS datasets across five sites. The talk closes with practical guidance for running StarPhase via a WDL workflow with standard inputs (reference genome, phased VCF, aligned BAM). This video is part of the Spotlight on Human Genomics series focused on practical tools and study-to-insight workflows.
00:00 Introduction and what is PGx diplotyping
02:34 Why long-read sequencing for PGx
05:10 StarPhase overview and gene coverage
06:10 Three algorithms: simple, HLA, CYP2D6
07:31 Visualization: CYP2D6 *4x2/*5 example
09:46 Benchmarking against prior outputs
11:31 Population analyses across ancestries
13:55 Novel alleles and rearrangements
14:55 How to run StarPhase (GitHub, Bioconda)
15:32 Closing
*Subscribe for more PacBio genomics content:*
youtube.com/@PacificBiosciences?sub_confirmation=1
Learn more about PacBio at pacb.com
Legal & Trademarks: Visit pacb.com/legal-and-trademarks
#pharmacogenomics #longreadsequencing #HiFiSequencing
