I am a Researcher

How can PharmGKB help me?

PharmGKB contains a variety of aggregated data and relationship summaries that can help you design and answer research questions related to pharmacogenomics. Curators at PharmGKB read primary literature and input each of the statistical relationships of a single genetic variant or haplotype and drug response into the database as variant annotations. These variant annotations are then aggregated so that all annotations on a single variant-drug combination can be found together in a clinical annotation. These clinical annotations are graded on a scale of 1-4, where a 1 and 2 represents relationships that are strong and consistent. You can also find illustrations of drug pathways, which illustrate which gene combinations are involved in the pharmacokinetics and pharmacodynamics of response to that drug, and very important pharmacogene summaries, which describe how variation in that gene affects drug response. Besides finding big picture summaries of genes and drugs, you can find details on specific genes, drugs, and variants. You can browse individual pages, or you can download all of the associations and relationships that are found in PharmGKB. Through these portals, PharmGKB can help you identify variables to include in your research on drug response, whether looking at specific genes or specific drugs, or a combination of the two. It can also help you interpret your findings in the context of what has been found by other research groups related to drug efficacy, dose, and toxicity. PharmGKB draws data from dbSNP and Drug Bank, and aligns our SNP definitions with a specific human genome assembly to aid compatibility with different data sources.

What are the builds/resources PharmGKB annotations are based on?

The resources that PharmGKB references and the versions currently used to map data on PharmGKB can be found here. These resources include dbSNP, a specific Human genome assembly, NCBI Gene, HapMap, and DrugBank.

Where can I find the search index and tutorials to use PharmGKB effectively?

You can learn more about how to use PharmGKB, what types of data can be found, and how to navigate the resources on the tutorials page. You can search for what information PharmGKB has on a specific gene, specific drug, or combinations of genes and drugs in the search bar.

Pharmacogenomic data terms

Interpreting pharmacogenetic variants can be complicated because of the range of common variability that is found in the genes. In addition, naming conventions are different than other disciplines in human genetics. Here are a few terms that will help orient you to the terminology conventions of pharmacogenetics.

  • rs number: each single nucleotide polymorphism (SNP) that has been submitted to dnSNP has a unique rs number. These rs numbers allow consistency in reporting variants, as they have been linked to a DNA base change at a specific genomic position. Some rs numbers also uniquely identify deletions or insertions.

  • Haplotype: a haplotype is a combination of variants that compose an allele. Unlike a SNP that refers to one position in the genome, a haplotype refers to the overall composition of a gene. This means that different haplotypes of the same gene can share some SNPs. The classic example of pharmacogenomic haplotypes is in the gene CYP2D6. CYP2D6 has over 100 haplotypes, many of which share SNPs and which include both gene duplications and deletions. Phenotypes are assigned to haplotypes rather than to specific SNPs in cases where haplotypes contain more than one SNP.

  • Star allele: traditionally, alleles in pharmacogenes have been identified through the "star nomenclature". PharmVar defines these star alleles for all cytochrome P450 genes. They link to the rs numbers and genome positions that define each allele. While rs numbers always refer to the single position on the genome, star allele definitions may vary slightly by research group. As a result, it is important to confirm the definition of the star allele in a given paper, whether reading or writing a paper.

  • Activity analysis: because of the vast number of variants that are possible in a given gene, alleles are classified as to their impact on protein function. Depending on the complexity of the genetic variation and the role of the protein in drug response, these phenotypes are described differently. Patient status with respect to the activity of a give gene reflect the combination of both of their alleles. CPIC has published a guide to the standardized terms here. CYP2D6 variants are given activity scores depending on the functionality of the protein produced by that allele. The activity scores of CYP2D6 alleles are defined.