The soaring costs of therapeutic research, estimated in the billions, result from frequent clinical trial failures, largely due to inefficacy. Studies reveal medicines backed by Mendelian loci and genome-wide association studies are two times as likely to acquire approval.
Despite keen interest in leveraging human genetics, present medication prioritization frameworks lack extensive integration of direction of effect (DOE) and various genetic data.
To bridge this gap, the authors present a Genetic Priority Score (GPS) for 19,365 genes and 399 medication indications, demonstrating its effectiveness in prioritizing targets and indicating clinical trial progress. Publicly available results feature a directional version, GPS with DEO (GPS-DOE), delivering enhanced understanding.
Genetic Priority Score Created to Prioritize Drug Targets and Indicate Clinical Trial Success
GPS was created to improve medication target prioritization founded on a dataset containing genes and medication indications. Employing phenotype-specific traits from clinical variants, coding variants, and genome-wide association loci, the authors incorporate data from different sources, including ClinVar, HGMD, UK Biobank, and Open Targets. After quality control, the resulting dataset incorporated 1,213 medications, 778 genes, and 282 medication indications for the discovery phase. Validation involved utilizing the SIDER 4.1 resource, delivering a dataset with 733 medications, 658 genes, and 325 medication indications. Univariate associations of eight genetic traits with drug indications show substantial correlations.
Publicly Released GPS-DOE Provides Enhanced Understanding of Genetic Mechanisms
The study also included conducting simulated experiments. Simulations utilizing the SIDER dataset show that high GPS values correspond to greatly improved chances of a gene being a drug indication. The study also observes a correlation between increased GPS values and medication advancement through clinical trial stages, highlighting the GPS’s capability to indicate successful medication targets.
Likewise, a directional version, GPS-DOE, is introduced to integrate the direction of genetic effect and drug mechanism, showing its association with drug indications. The GPS is then expanded to 19,365 genes and 399 phenotypes, pinpointing likely new drug targets and informing various phases of medication development.
GPS Pinpoints Promising Therapeutic Targets Across Drug Development Pipeline
In summary, GPS was created to improve the prioritization of therapeutic targets in medication discovery. The GPS incorporates numerous phenotype-specific genetic traits, delivering a comprehensive method for pinpointing successful medication targets. Despite certain impediments, such as discrepancies in medication ascertainment between datasets and possible misclassifications, the GPS demonstrates value in proposing likely therapeutic targets across different phases of medication development, from preclinical to advanced clinical trials.
