Developing Standards for Genetic Variation Datasets from Emerging Technologies

1
Developing Standards for Genetic Variation
Datasets from Emerging Technologies

• Tina Hernandez-Boussard
• Stanford University, PharmGKB
• boussard_at_stanford.edu

2
Outline of Talk

• Background
• Understanding genetic variation
• PharmGKB
• Emerging Technologies
• High density SNP arrays
• Studies generating using SNP arrays
• Standards
• FuGO
• PML
• Challenges

3
Genetic Variants

• Today, renewed emphasis on population variation,
  exemplified HapMap Project
• The exploration of quantitative variation in
  human populations has become one of the major
  priorities for medical genetics.
• Systematic mapping of gene-association
  susceptibility genes for complex diseases is
  underway.
• The successful identification of variants that
  contribute to complex traits and adverse drug
  reactions is highly dependent on reliable assays
  and these genetic maps.

4
Where are we in 2006?

• The human genome has been sequenced
• Variation in the genome has been characterized
• We are working on developing a comprehensive map
  of human genetic variation to guide efficient
  marker selection HapMap
• We can measure variation in phenotypes at all
  levels
• Molecular, cellular, organ, and organism
• Technology has been developed
• High-throughput genotyping platforms
• We are learning about the consequences of genetic
  variation

5
Pharmacogenetics Pharmacogenomics

• Understanding how genetic variation leads to
  variation in responses to drugs
• A promise from the Genome Project
• Pharmocogenomics a genome-wide study of
  pharmacogenetics
• Personalized Medicine
• Making drug use effective and safe based on a
  persons specific genotype

6
PharmGKB

• A public genotype-phenotype research database
• Disseminate public data sets about genomic,
  molecular and clinical variability in disease and
  in response to medications.

7
Types of data in PharmGKB

• Genotype
• Primary data
• Transporter variation
• Phenotype
• Varied and individual
• Literature Annotations
• Added by scientists and researchers
• Pathways
• Developed by hand
• Supported literature annotation

8
Knowledge and Data in PharmGKB
9
PharmGKB Users

• Geneticists (Scientists seeking access to
  polymorphism data or phenotype data about
  variants)
• Experimental Biologists
• Pharmacologists/clinical scientists with focused
  questions about how a drug may be metabolized
• Outcomes researchers
• General public with interest in science
• Eventually, clinical information systems but NOT
  NOW

10
Simple Use Cases

• For Gene X, show all polymorphisms in sequence
  that have been observed.
• For Drug Y, show variability in pharmacodynamics
• For Phenotype Z, show variability in association
  with drug or genetic allele
• Download database contents with polymorphism and
  phenotype associations

11
PharmGKB HomePage
12
PharmGKB GenePage
13
PharmGKB Variant Page
14
PharmGKB Frequency Data
15
PharmGKB Variant Page
16
PharmGKB Variant Information
17
PharmGKB GenePage
18
PharmGKB Phenotype Data
19
PharmGKB Pathways
20
Emerging High-Throughput Technologies

• As high-throughput genomic technologies become
  more relevant in biomedical research, PharmGKB is
  adding functionality to accommodate these assays
• High-throughput SNP arrays
• Microarray data
• Protein arrays
• Tissue arrays
• More drug-response data
• Metabolomics

21
Genome-wide SNP Arrays

• Genotype thousands of single nucleotide
  polymorphisms in a single assay
• Ability to revolutionize the ability to
  identify disease-associated proteins and their
  corresponding pathways as drug targets
• Studies using this technology include
• Linkage studies
• LOH CGH studies
• Association studies

22
Linkage studies

• Density 3,000 to 10,000 markers
• Biallelic
• Evenly distributed throughout genome
• Successfully used to track genomic regions as
  they co-segregate with a disease through a
  pedigree.
• Benefits when compared with traditional
  microsatellite markers
• Time saved
• Cost
• overall information content increased by 20
• false-positive rate of 0.05 - 0.08

23
LOH and CGH studies

• 10,000 to 100,000 markers
• LOH Studies based on the model that one inherited
  allele is mutated and the other is lost
  somatically
• CGH Studies show whole regions of the chromosome
  that are amplified or deleted
• carry candidate oncogenes or potential tumor
  suppressor genes
• Significantly enhanced our ability to detect
  chromosomal aberrations in cancer cells and
  assess their role in tumorigenesis

24
Association Studies

• 100,000 to 500,000 markers
• Most diseases are influenced by genetic and
  environmental factors complex diseases.
• To study these diseases a large sample of
  well-placed markers throughout the genome
• Markers on the array are strategically placed and
  expected to fall within close proximity to true
  disease causing variants.
• Markers are in linkage disequilibrium with a
  disease-causing mutant.

25
Genotyping Technologies Supported by PharmGKB

• Array Technologies using hybridization
• Affymetrix GeneChip
• Illumina BeadArray
• Mass-Spectometry Technology
• Sequenom MassARRAY

26
What should be reported?

• rs and/or the ss
• Location given by chromosome and coordinate
• Strand
• Call
• SNP call from the strand reported
• Raw intensity values
• Normalized intensity values
• Score given by software

27
Standards

• FUG-O (Functional Genomics Ontology) will be used
  to capture array designs and experiment
  annotations
• being developed to model functional genomic
  domains such as Proteomics, Metabolomics, as well
  as Transcriptomics
• Use PML (Polymorphism Mark-up Language) to
  distribute data and encode annotations about SNPs
• OMG approved standard format for SNP information

28
Functional Genomics Ontology - FuGO

• Goals
• Unambiguous description of how the investigation
  was performed
• Consistent annotation, powerful queries and data
  integration
• Participants technological biological
• HUPO-PSI, MGED, Metabolomics
• RSBI working groups
• Polymorphism community is working with FuGO to
  ensure communitys interests are met

29
FuGO Goals

• Provide descriptors for experimental components
• NOT model biology NOR the experimental workflow
• Universal descriptors
• Biological and technological domain-specific
  terms
• Sharing of software and tools with existing
  bio-ontology

30
PML

• Polymorphism MarkUp Language
• XML-based format for genotype data exchange
• International Consortium
• Platform Independent Model
• Platform Dependent Model
• dbSNP
• HapMap
• HGVbase
• PharmGKB,

31
PML Use

• Use cases
• Data upload or download between databases
• an exchange of individual data items (subset)
• Ontology support (next version)
• http//pml.ddbj.nig.ac.jp/index.html

32
Conclusions

• Understanding genetic variants will lead to a
  better understanding of disease and drug
  variation
• PharmGKB is building public database to catalyze
  genotype-phenotype research and deployment.
• Post-genome informatics challenges require
  marriage of bioinformatics and clinical
  informatics
• Data standards have to be developed and sustained
  to meet our challenges
• Pharmacogenetics holds promise as a form of
  individualized medicine, but it is not yet
  routine.

33
Thank you to Helen Parkinson, Prof. Begent, the
Organizing Committee for this invitation

• On behalf of the PharmGKB team for providing us
  an opportunity to engage the world-wide
  scientific community in this project by talking
  about our work.