Transcript Hons Pharmaco DBsx

Pharmacogenomics
databases
Nicky Mulder
Some info extracted from slides from Russ Altman’s pharmacogenetics course
What do we need?
• Need machine learning and other tools for associating genotypes to
phenotypes
• Databases with useful info but also adequate security and privacy
• Database of all known PK and PD genes
• Knowledgebase of drugs, gene-drug interactions, dosing info, etc.
• Tools for studying effects of SNPs
• Tools to enable point of care usage
Informatics issues in PG
• Standards for data representation & exchange
• Genes
• Phenotype
• High-throughput data
• Analysis of literature
• Find pgx papers
• Extract gene-drug interactions
• Integration of data from multiple sources
• Gene sequence, protein sequence
• SNP databases
• Drug info databases
• Representation and manipulation of pathways (PK & PD pathways)
Standard terminologies for PGx
• Need standard terminology/ontology for concepts:
• Genes
• Pathways
• Drugs
• Diseases
• Symptoms (Adverse events)
From PharmGKB
Ontologies
• Formal specification of terms with relationships between them
• Usually organised in hierarchical structure
• Terms should be defined
• Allows unambigiuous naming of entities
• Facilitates searching and data exchange
Example ontologies
• Gene Ontology (gene products)
• Disease ontology
• Phenotype ontology
• Anatomy ontology
• MESH terms
• SNOMED
Disease ontology
• Developed to create a single structure for the classification of disease
which unifies representation of disease among the many varied
terminologies and vocabularies into a relational ontology
• Originally (2003, 2004) based on ICD-9 code ICD = International Statistical
Classification of Diseases and Related Health Problems developed by WHO
• Organized into eight main nodes:
•
•
•
•
•
•
Cellular proliferations e.g. cancer
Mental health
Anatomical entity
Infectious agent e.g. anthrax
Metabolism
Genetic diseases
DO example
PATO: Phenotype and Trait Ontology
• Generic ontology for describing traits or phenotypes that can be
measured either quantitatively or qualitatively (e.g. bone length).
• Species independent
• Can be used in combination with a wide range of other ontologies
• Consists of two main term types:
• Attributes
• Values
• These are related to each other
Human phenotype ontology
• http://www.human-phenotype-ontology.org/
• standardized vocabulary of phenotypic abnormalities encountered in
human disease
• Developed using the medical literature, Orphanet, DECIPHER, and
OMIM
• Contains ~11,000 terms and >115,000 annotations to hereditary
diseases
Phenotype example
SNOMED
• http://www.nlm.nih.gov/research/umls/Snomed/snomed_main.html
• Systematized Nomenclature of Medicine ->Clinical Terms
• Arranged in a hierarchy with levels of specificity
• Used in electronic health records
SNOMED
structure
http://ihtsdo.org/fileadmin/user_upload/doc/download/doc_StarterGuide_Current-en-US_INT_20141202.pdf?ok
Standard terminologies for PGx
• Need standard terminology/ontology for concepts:
• Genes
• Pathways
• Drugs
• Diseases
• Symptoms (Adverse events)
From PharmGKB
Genes and variants
• Genes:
• Human Genome Nomenclature Committee (http://www.genenames.org/)
• GeneCards (http://www.genecards.org)
• OMIM (http://www.omim.org/)
• SNPs and their locations
•
•
•
•
•
dbSNP
HapMap
Human Genome Mutation Database
HGVBase
Ensembl, UCSC browser
HGNC
GeneCards
Variant data GeneCards
Variant data dbSNP
Pathways
• MetaCYC
• KEGG
• Reactome
• Biocarta
• And many others (including pathway display programs cytoscape,
genmapp)
Pathways from GeneCards
Drug data
• KEGG drug taxonomy: http://www.genome.jp/kegg/drug/
• DrugBank
• PubChem
• MedlinePlus
• RxNorm
• PHARMGKB includes all of these
KEGG drug taxonomy
• Has target-based classification:
•
•
•
•
•
•
Antiinfectives
Antineoplastics
Antidiabetic agents
Neuropsychiatric agents
Anti-allergic agents
Topical steroids
• Has drug interaction database with known adverse drug-drug
interactions associated with contraindications (CI) and precautions (P)
• Drug-structure maps –development through chemical structure
KEGG search for aspirin
Aspirin page
Interactions with
other drugs
Pharmacogonetics/genomics databases
• Aim is to gather data on pharmaco-genes and pathways with links to
relevant sources, e.g. drugs, gene info, variant info, etc.
• Examples:
• PharmaADME
• ePKGene
• PharmGKB
PharmaADME
• http://www.pharmaadme.org/joomla/
• List of drug metabolizing (ADME) genetic biomarkers applicable to
pharmaceutical clinical trials & FDA
• Developed by industry
• Provides list of genes “to be screened to identify predictors of
pharmacokinetic variability that could impact drug safety and efficacy in
the current drug development process”
• Genes grouped into four categories:
• Phase I and II metabolism enzymes, responsible for the modification of functional
groups and the conjugation with endogenous moieties respectively;
• transporters, responsible for the uptake and excretion of drugs in and out of cells;
• modifiers, that can either alter the expression of other ADME genes or affect the
biochemistry of ADME enzymes
ePKGene
• https://www.druginteractioninfo.org/applications/pharmacogeneticsdatabase/
• Pharmacogenetics DB from U. Washington
• Can search for a compound –links to PubChem
• Provide gene(s) whose SNPs may affect response to drug
• Provides info on populations
• Can search by gene and get variants and their impact
PharmGKB
• https://www.pharmgkb.org/
• “PharmGKB is a comprehensive resource that curates knowledge
about the impact of genetic variation on drug response for clinicians
and researchers. ”
• Can search genes, drugs, diseases and pathways
• Includes variant annotations, drug-centred pathway, clinical info, drug
dosage guidelines, gene-drug associations, etc
• Uses ontologies
What PharmGKB does
• Annotates genetic variants and gene-drug-disease relationships from
literature
• Summarizes important pharmacogenomic genes, associations between
genetic variants and drugs, and drug pathways
• Curates FDA drug labels containing pharmacogenomic information
• Curates and participates in writing pharmacogenomic-based drug dosing
guidelines
• Publishes pharmacogenomic-based drug dosing guidelines, important
pharmacogene summaries and drug-centered pathways
• Displays all information on the website and provides comprehensive
downloads
https://www.pharmgkb.org/page/overview
PharmGKB link between genes and PK/PD
https://www.pharmgkb.org/page/overview
PharmGKB home page
A tour through PharmGKB
Search for CYP2D6
Downloads
available from
PharmGKB
Summary
• Pharmaco-related data includes:
•
•
•
•
Genes, functions, variants
Pathway
Disease, phenotype, adverse effect
Drugs
• May be multiple entry points
• Essential to use standard terminologies and ontologies
• Aggregated into pharmacogenomics databases
Personalized/precision medicine
• Medicine is personal:
• We are all different
• Some of our differences translate into how we react to drugs as individuals
• Therefore personalized medicine is important
• Issues:
•
•
•
•
Some people need twice the standard dose to be effective
Drugs may work for some but not others
Some people have side-effects others don’t
Some people get diseases and others don’t
• Goal: The Right Dose of The Right Drug for The Right Indication for
The Right Patient at The Right Time.