Transcript No Slide Title

Genetics in Clinical Research
Jonathan L. Haines, Ph.D.
Center for Human Genetics Research
7/16/04
CLASSES OF HUMAN
GENETIC DISEASE
• Diseases of Simple Genetic Architecture
–
–
–
–
Can tell how trait is passed in a family: follows a recognizable pattern
One gene per family
Often called Mendelian disease
“Causative” gene
• Diseases of Complex Genetic Architecture
– No clear pattern of inheritance
– Moderate to strong evidence of being inherited
– May be:
• common in population: dementia, stroke, tremor, etc.
• Rare in population: Adverse drug response, primary lateral sclerosis, etc.
– Involves many genes or genes and environment
– “Susceptibility” genes
– This is your trait!
COMMON COMPLEX DISEASE
Environment
Epidemiologists
Genotyp
e
Complex Disease
Geneticists
Phenotyp
e
Clinicians
Analysis
Biostatisticians
Why Test the Genes?
• Basic Science: Better understanding of biology
of disease
– Direct probe into functional pathways
– Target for detecting interacting factors
– Better definition of disease
• Clinical Science: Making this knowledge useful
–
–
–
–
Improved diagnostic testing
Presymptomatic testing
Improved prognostic testing
Improved treatment (e.g. pharmacogenetics)
Why test the Genes?
Practical reasons
• Can help make sense of results
– If there is a lot of variability, it may be due to
genetics
– Can clean up the analysis and find significant
results!
– Can add a sexy new component to your study
– It can be easy and cheap through the GCRC!
• Virtually all GCRC studies have a potential
genetic component
• Pilot data can lead to larger funded studies
DISEASE GENE
DISCOVERY
• Broad Search (Genomic screen)
– Examine a large but representative subset of all genomic
variations. Not hindered by poor assumptions of biology.
– Use families with more than one affected individual.
– Problem: Lots of genes at the same location!
• Targeted Search (Candidate genes)
– Examine a specific and small set of candidate variations
based on what we know about the biology of the disease.
– Can use both families with multiple affected individuals and
families with only one affected individual.
– Problem: There are 50,000 genes and we know very little
about their function!
Genome Project
Genome Toolbox
• Physical map: genome
sequence
– multiple different species
• Genetic map:
recombination
• Linkage disequilibrium
map: HapMap
• Variation maps
–
–
–
–
Repeats
Deletions
Duplications
SNPs
• Homology maps
• Hardware Technology
– Sequencing
– Genotyping
• Software Technology
– Public databases
– Analysis programs
• Increased productivity
– Experiments now possible
that were considered
impossible just 2 years ago
Study
Study Designs
Designs
Linkage Analysis
Large Families
Small Families
Association Studies
Family-Based
Case-Control
Association Study Designs
• Family-based analysis
– Two flavors
• Trio (patient and both parents)
• Discordant sibpairs
– Multiple statistical methods for analysis
– Advantage: inherent control for genetic background
– Disadvantage: family-based
• Case-Control
– Standard epidemiological design
– Statistical methods
• logistic or linear regression
• Statistical genetics methods
• Case only
– Outcomes analysis
Genetic Association Analysis
• Can incorporate gene/gene interactions
– Look at two or more genes at a time
• Logistic regression
• MDR
• Can incorporate gene/environment
interactions
– Logistic regression
– MDR
Need To Characterize The Gene
• Use genome databases to get known
information
–
–
–
–
–
Gene location (NCBI, Ensembl, Celera)
Gene structure (NCBI, Ensembl, Celera)
Possible gene functions (OMIM, NCBI, KEGG)
Gene expression (tissue localization) (NCBI)
Gene variation (HapMap, dbSNP, Celera, OMIM)
•
•
•
•
Deletions
Mutations
SNPs
LD relationships
Gene Characterization
• Choose what variants to examine
• Decide if further polymorphism
discovery is needed
• Many factors to be considered:
–
–
–
–
Frequency of variant
Genotyping platform and assay development
LD relationships
Availability and quality of DNA
Essential Problem in Choosing
the Gene(s) to Study
• How do we integrate all the available
information that we and others generate?
• How do we locate the one or few genetic
variations involved in our trait in the sea of
hundreds or thousands of possible variations?
• Most methods identify a set, often a large set, of
possible variations.
Genomic Convergence
Metabolism
Association
Expression
Genomic convergence identifies the intersection
of genes found through multiple methods such as
drug metabolism, allelic association analysis, and
gene expression studies.
Using Your Time and Effort Wisely
• Design your study. Genetics can be
added easily and will only benefit, not
hinder, the main study
• Do not waste time on the details! We
have the expertise to help make it
happen.
Core Services
(http://chgr.mc.vanderbilt.edu/core.shtml)
• Family
Ascertainment Core
– 1207 17th Ave S., Suite
100
– Kelly Taylor, MS;
Manager
• DNA Resources Core
– 518 Light Hall
– Cara Sutcliffe, MS;
Manager
• Genetic Data
Analysis Core
– 511-515 Light Hall
– Chun Li, Ph.D.;
Faculty Advisor
• Computing/
Bioinformatics Core
– 515-519 Light Hall
– Janey Wang, MS2;
Manager
Family Ascertainment Core
Faculty advisor: Jeff Canter
Kelly Taylor
Manager
Genetic Counselor
Genetic Counselors
Research Nurses
Research Associates
Amy Crunk
Benita Lynch
Regan Bergenslasser
Molly Klein
Lynne McFarland
Clare Knebush
Geana Crockett
Denise Fuzzell
Shelly Stocki
Student Workers
Family Ascertainment Core Services
•
•
IRB and Protocol Development
Patient/Family Ascertainment
– Identify and recruit participants
• Clinic, local, distant
– Data collection
• Family history, clinical, demographic
– Biological sample collection
• Phlebotomy
• Buccal washes
• Finger sticks
•
Project and Data Management
– Progeny pedigree and ascertainment database
– PEDIGENE clinical and genetic database
– Template forms
• IRB
• Family History
• Clinical
– Limited access, locked file room
DNA Resources Core
Faculty Advisor: Doug Mortlock
Cara Sutcliffe
Manager
Ping Mayo
Elizabeth Mathews
Maria Comers
Student Worker
DNA Resources Core
Services
• DNA extraction
– Blood
– Buccal (wash, brush)
– Cell Pellets
• Sample tracking and
storage
– Web-based Oracle
database
– PI-controlled access
– Bar-coded, standardized
storage in locked cold
room
• DNA quantitation
• Initiation of lymphoblast
cell lines
• Microsatellite
genotyping
• SNP genotyping
• Storage of cell lines (LN2
freezers)
DNA Resources Core
Resources
• 4 Staff
• Automated DNA large and small volume
extraction (Autopure)
• Locked cold room, liquid nitrogen freezers
• Bar-coding, RPIDs, web-based database
• Hitachi FMBIO II laser scanner (fluorescent
dyes)
• ABI 7900HT (high-throughput SNP
genotyping)
Genetic Data Analysis Core
Faculty Advisor: Chun Li
Chun Li
Managing Director
Data Analysts
Data Management
Lana Olson
Valerie Mitchell
Lan Jiang
Kim Gainer
Yuki Bradford
Connie Qualls
Jackie Bartlett
Genetic Data Analysis Core
Services
• Linkage Analysis
– Parametric lod scores Nonparametric scores
– Two disease loci
• Linkage Disequilibrium
Analysis
– Case-control
– Family-based (TDT, S-TDT,
PDT)
• Gene-Gene Interactions
• Gene-Environment
Interactions
– Logistic Regressions
– MDR analysis
• Quantitative Trait Locus
Analysis
• Marker reference maps
• Error detection
– Mendelian checks
– Haplotype checking
– Pedigree relationship
checking
• Consultation on study design
• Training on use of software
• Data Management
Genetic Data Analysis Core
Resources
• 7 Staff
• PEDIGENE database
– Clinical
– Family history
– Genotyping
• Latest genetic analysis
software
– Testing of new programs
and methods
– Experience with strengths
and weaknesses
• Access to
–
–
–
–
7 PCs
6 Unix
12 Linux systems
VAMPIRE
Computing/Bioinformatics Core
Faculty advisor: Marylyn Ritchie
Janey Wang
Manager
Database Support
Database Support
Bioinformatics Support
Doug Selph
Database Administration
Sunny Wang
Programming
Lixin Chen
Database Programmer
Christian Shaffer
Programming/Web Design
Huiming Li
Database Programmer
Nate Barney
Programming
Computing/Bioinformatics Core Services
• Complete database services
– Support of PEDIGENE
– Data collection
• Web-based data entry
• Teleforms scannable forms
– Oracle expertise
– Build custom databases
– Extend current databases
• Bioinformatics Support
– Programming/scripting
– Web-design
Getting your GCRC Genetic Study
Done
• Develop Protocol:
– New protocol: Contact Kelly Taylor, she’ll do most of the work
– Existing protocol: In many cases existing DNA addendum will work
• Present to GCRC for approval
– <100 DNAs, <100 genotypes, no special approval needed
– >100 DNAs, >100 genotypes, Genetics subcommittee must approve.
• Perform study
– DNA collection can be done on GCRC or by FAC
– DNA extraction by DNA Resources core
– Genotyping by DNA Resources core
• Genetic Analyses
– Can be done by DMAC (additional fee)