Instructions for Viewers • To share webinar via social media: • To see speaker biographies, click: View Bio under speaker name • To ask a question, click the Ask A Question button under the slide window • To share webinar via e‐mail: Advances in addiction research Applying genetic biomarkers to personalize treatments June 3, 2015 Webinar Series Sponsored by:
58
Embed
Instructions for Viewers - Science · Instructions for Viewers • To share webinar via social media: • To see speaker biographies, click: View Biounder speaker name • To ask
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Instructions for Viewers
• To share webinar via social media:
• To see speaker biographies, click: View Bio under speaker name
• To ask a question, click the Ask A Question button under the slide window
• To share webinar via e‐mail:
Advances in addiction researchApplying genetic biomarkers to personalize treatments
June 3, 2015
Webinar Series
Sponsored by:
Participating ExpertsBrought to you by the Science/AAAS Custom Publishing Office Andy Brooks, Ph.D.
Rutgers UniversityNew Brunswick, NJ
James Baurley, Ph.D.BioRealm, LLCMonument, CO
Advances in addiction researchApplying genetic biomarkers to personalize treatments
June 3, 2015
Webinar Series
Sponsored by:
Precision Medicine: A Scientists Guide to Transformational Technologies and Novel ApplicationsDr. Andrew BrooksCOO, RUCDR Infinite BiologicsScientific Director, Bioprocessing Solutions Alliance
3
6/3/2015 4CONFIDENTIAL
• Integrating sample collection, bioprocessing and management with an eye on the molecular central dogma
• Understanding the power of evolving technologies and developing a sample centric roadmap for future sample use
• Creation of sample quality control metrics to standardize across all collections
• Creating a resource that will integrate seamlessly with both industrial and academic collaborative opportunity
Science and Technology:Driving Innovation
• Diagnostics, Therapeutics and Pharmaceuticals tailored for “you” in your specific environment
Defining Importance of Personalized/Precision Medicine
5
Disease Classification: Shift from Anatomy/Histology to Molecular Etiology
6
• Biology Redefined
Lung CancerNon‐Small Cell
MacConaill L E , Garraway L A JCO 2010;28:5219-5228
Sample Collection Errors
2%98%
Percentage of misidentified samples in a biobank
Percentage of collection errors that occur outside of the biobank
Dr. Andy Brooks in “Q&A: RUCDR's Andy Brooks on the Challenges Facing Biorepositories and the Rise of Biobank Arrays”BioArray News, April 23, 2013
Management of complete lifecycle andcomprehensive sample management is critical
Manage the Sample Lifecycle
9
STORAGE
EXTRACTION
COLLECTION
Potential for Adverse Sample Conditions
Limiting Biological Resources:Growing Demand for Access
• Best sample preservation/ processing practices of yesterday may not be best choice for future molecular analyses
• Use least amount of possible primary sample for analysis
• Goal is to better integrate sample collection directly with biosample processinglaboratory activities
Illustrates decreasing size of samplecollections and increasing distributionof samples to sites for analysis byBioprocessing Solutions Alliance
• Maximal use of primary samples– Undefined application for downstream analyses
• Efficient processing– Maximizing extraction technologies to improve yield and
quality• Appropriate storage
– Defining storage formats and temperatures to maximize storage infrastructure
• Nucleic acid amplification / Cell line establishment– Creating renewable resources to preserve primary sample
and/or precious collections• Appropriate distribution guidelines
– Define needs for specific downstream applications to preserve sample resources
DNA Extraction/QC Workflow Evolution
OLD
NEW
GenderPolymorphic
13
Define the affects of storage Create processing and aliquoting strategies
that are sample dependent Create robust biomaterials that can be
compared across sites (i.e. gene expression analysis)
Understand the Limitations and Utilization of Sample Types
14
Comprehensive Solutions for Tissue Sample Preservation and Processing
Nucleic Acid Quality Application Performance
Variability (expression) Histology/IHC
% S
ucce
ssfu
l Sam
ples
Qua
lity
Sco
re%
Con
cord
ance
Qua
lity
Sco
re
Frozen Tissue FFPE RNALater PaxTissue
Evaluate sample quality as a function of downstream applications
Understand laboratory impact of sample variability
Preserve both clinical and research analytical and diagnostic workflows
Create functional QC metrics for every sample type
6/3/2015 15
• Increasing need for more data sooner to support translational strategies
• Growth in “omics” technologies is increasing the diversity of sample types and sample analytesthat need to be managed
• Improved visibility to better annotated samples to support sample selection, research utilization and prevention of duplication
• Movement towards companion diagnostics is increasing need for sample collection and storage
• Understanding of the quality of specimens is vital so the right samples are stored and selected for downstream processing and analytical testing
CONFIDENTIAL
Science is a Key Driver for Precision Medicine Program Development
Growth in targeted drug therapies and personalized medicine, combined with the increasing pressure on biopharmaceutical companies to advance research in a more cost efficient manner is increasing the demand for high‐quality, well‐annotated biological samples.
Bioprocessing Solutions AllianceInnovative Service Solutions
• Earlier baseline genetic risk profiling• Sequencing genomes at birth
• Real time health/genomic monitoring• Biosensor arrays, Smart Homes
• Personalized Rx solutions• Comprehensive medical, genomic and environmental data integration
Future of Personalized Medicine
21
• Global leader in comprehensive sample management solutions
– Prepares and stores millions of samples for bioscience R&D
• Pioneer of Good Storage Practices (GSP)• Offer a state‐of‐the‐art approach to
sample preparation and renewable resourcing of research samples
• Provides innovative ISISS® sample management technology system that provides web‐based access to global sample inventories
• Sample logistics and relocation expertise
• World’s largest genetics based biorepository and top tier research center
– Distributes over 1 million biological and cell‐based samples for discovery research
– Customers include: Mayo Clinic, Sloan Kettering Cancer Center, Army, NIH
• Provides the highest quality biomaterials, technical consultation and logistics
• Pioneers innovative technology platforms– Stem cell research and nucleic acid QC and
nanofluidics technologies
• Utilizes proprietary products such as SNPtraceTM to perform functional quality control
Bioprocessing Solutions Creates a Standardized & Global Operational Environment
22
BiobankingRedefined
6/3/2015 23
• Upstream clinical sample data elements extracted and sent electronically• Sample registered in ISISS® sample management inventory database • Samples are stored and mined for downstream bioprocessing• BSALIMS captures laboratory testing data • Sample inventory data connects to BSALIMS data to improve future sample selection upstream
CONFIDENTIAL
BioProcessing & Biobanking Integration: Sample and Data Harmonization
Participating ExpertsBrought to you by the Science/AAAS Custom Publishing Office Andy Brooks, Ph.D.
Rutgers UniversityNew Brunswick, NJ
James Baurley, Ph.D.BioRealm, LLCMonument, CO
Advances in addiction researchApplying genetic biomarkers to personalize treatments
June 3, 2015
Webinar Series
Sponsored by:
Smokescreen® ‐ a unified platform for addiction research
BioRealm, LLC
James Baurley, Ph.D.Co‐founder and Principal Data Scientist
27
Motivation
Motivation
Individual responses are influenced by variation in biological pathways
30PharmGKB nicotine metabolism pathway. Image by PharmGKB and Stanford University.
Motivating Questions
• How does variation in G, E, pathways influence…– Dependence?– Treatment response?– Risk of related diseases?– Adverse events?
31
A uniform platform is needed to...
• Capture genotypes
• Combine it with phenotypes and exposures
• Analyze data in new ways
32
Our requirements
• Common panel• Hypothesis testing and discovery• Fully customizable array• Reproducible manufacturing• Automated and high quality workflows• Paired software and services for common bioinformatics and analysis tasks
• Cost
33
Patient biospecimen collection
Raw data uploaded for processing
De-identified data entered into secure
web application
DNA samples processed at partner lab
Algorithms applied to cleaned dataReports
A uniform and simple process
296K Imputation‐based GWAS backbone (multi‐population) 11K Fine‐mapping markers for
CHRNA5‐CHRNA3‐CHRNB4 and CYP2A6‐B6
255K255KMulti‐population tag SNPs for >1000 addiction‐related genes17K Exome SNPs and indels5K Loss‐of function markers
5K Top NeuroSNP markers2K PNAT SNPs12K Quit Success Score v1.0 SNPs657 High‐priority literature SNPs1.9K Tobacco smoke constituent uptake and metabolic phenotype markers
Smokescreen® Genotyping ArrayA platform for addiction research
9K UK Biobank lung function markers2K UK Biobank cardiovascular markers1.2K Psychiatric markers3K Lung cancer markers8K NHGRI GWAS Catalog markers
Comorbidities and DiseasesCandidate Addiction Genes
Genome‐wide Discovery Fine‐mapping
General Content
Priority Addiction Content
Affymetrix® Axiom® Platform>646K SNPs
35
Imputation‐based genome‐wide coverage
MAF≥ 0.05
MAF≥ 0.01
% SNPs imputation r2 ≥ 0.8
36
†Nelson SC, Doheny KF, Pugh EW, et al. Imputation‐based genomic coverage assessments of current human genotyping arrays. G3 (Bethesda). 2013;3(10):1795‐807.‡Data calculated by BioRealm using methods described in Nelson et al. (2013)
Candidate Addiction GenesNeuroSNP Project
Pharmacogenetics of Nicotine Addiction Treatment (PNAT)
Tennessee Mouse Genome Consortium
1000+ Addiction Genes• Pairwise tagging using 1000 Genomes data, June 2012 (255K tag SNPs)
• Includes rare exonic, LoF, and eQTL markers
• >97% coverage (MAF ≥ 0.05, r2 ≥ 0.9) in East Asian (ASN), European (EUR), and Africans (YRI)
9K UK Biobank lung function markers2K UK Biobank cardiovascular markers1.2K Psychiatric markers3K Lung cancer markers8K NHGRI GWAS Catalog markers
Comorbidities and DiseasesCandidate Addiction Genes
Genome‐wide Discovery Fine‐mapping
General Content
Priority Addiction Content
Affymetrix® Axiom® Platform>646K SNPs
45
Smokescreen® Analytic Software and ServicesA platform for addiction research
46
Smokescreen® Analytic Software and ServicesA platform for addiction research
Quality control of Smokescreen data
1 Axiom Best-Practices workflow
2 Check for batch effects
3 Genotyped sex check
4 Replicate sample concordance
5 SNP reproducibility in sample replicates and Mendelian error checks in trios
6 Hardy-Weinberg Equilibrium calculation
7 Identification of duplicates with different subject identifiers
8 Identification and/or verification of related individuals
9 Excess heterozygosity check
10 Sample and SNP quality statistics and reports
11 Analysis ready data delivery in PLINK format
12 dbGAP submission
47
Smokescreen® Analytic Software and ServicesA platform for addiction research
Analysis of Smokescreen Data
1 Ancestry estimation with Structure and fastStructure software packages
2 Principal component analysis (PCA)
3 Imputation to 1000 Genomes Project
4 Select samples and variants for genetic association scan
5 Combine results in meta-analyses
Interpretation and Reporting
1 Ancestry and population structure graphics
2 Reporting of clinically relevant panels (PharmGKB panel) and biomarkers
3 Tables and figures
48
Nicotine metabolism in multiple populations
• Individuals with higher rate of nicotine metabolism and clearance are more likely to have higher smoking exposure levels (Benowitz, et al. 2003).
• Understanding population‐specific variation in nicotine metabolic genes needed for biomarker development and optimization (Quaak, et al. 2009, Whirl‐Carrillo, et al. 2012).
• Study evaluated variation in 12 nicotine metabolic genes in relation to the nicotine metabolism ratio (NMR) of trans‐3’‐hydroxycotinine to cotinine in multiple populations.
• 346 individuals from three laboratory‐based studies of nicotine metabolism were genotyping using the Smokescreen platform.
49
In the liver, enzymes from cytochrome P450 (CYP) and UDP‐glucuronosyltransferases (UGT) families take nicotine as a substrate, influencing its metabolites’ ratios (Benowitz, et al. 2009).
50PharmGKB nicotine metabolism pathway. Image by PharmGKB and Stanford University.
Study Population
• Individuals selected from three laboratory‐based studies of nicotine metabolism:– Pharmacokinetics in Twins (“PKTWIN”, Swan et al., 2003)
– Pharmacogenetics Study of Nicotine Metabolism (“588”, Dempsey et al., 2004)
– Nicotine Metabolism Study on Individuals with smokers as relatived (“SMOFAM”, Swan et al., 2003)
• Analysis in EuA (n=160), AfA (n=48), AsA (n=52). Combined in meta‐analysis.
51
Nicotine metabolism analysis plan
52
P explanatory variables Yi NMR for individual iX matrix of P variablesE (Yi ) = μi
g link functionβp regression coefficient for variable pIp indicator if variable p is in the model
Discussion• CYP2A6 and UGT2B10 together are responsible for metabolizing and clearing 85‐90% of nicotine intake (Murphy et al. 2014)
• CYP2B6 metabolizes bupropion, used for cessation to reduce nicotine craving (Bloom et al. 2013).
• Association pattern and significance with NMR remained intact for CYP2B6 after analysis adjusted for CYP2A6’s top SNP (other CYP2A6 SNPs also retained significance)
• These markers may be used to model smoking behavior, treatment response, and tobacco‐related diseases.
54
EExposures
G
X1
Genes
Unobserved intermediate events
YDisease
“Topology” of the networkZ
External biological knowledge (“Ontologies”)
X2 X3
XnXn‐1
Next steps – a pathway perspective to prediction
Main messages
• Trend towards custom targeted arrays (AffymetrixAxiom) and paired analytic software and services to interpret data.
• Biomarker‐based study designs are powerful at characterizing associations in diverse populations.
• Select array content to match your research questions.• Multivariate statistical approaches are needed to develop prediction models.
• Cycle needed between research and numerous clinical products
56
Participating ExpertsBrought to you by the Science/AAAS Custom Publishing Office To submit your
questions, click theAsk a Question
button
Advances in addiction researchApplying genetic biomarkers to personalize treatments
June 3, 2015
Webinar Series
Sponsored by:
Andy Brooks, Ph.D.Rutgers UniversityNew Brunswick, NJ
James Baurley, Ph.D.BioRealm, LLCMonument, CO
For related information on this webinar topic, go to:
Look out for more webinars in the series at:
webinar.sciencemag.org
To provide feedback on this webinar, please e‐mailyour comments to [email protected]
Brought to you by the Science/AAAS Custom Publishing Office
Advances in addiction researchApplying genetic biomarkers to personalize treatments