SNP-analysis Linda Broer (l.broer@erasmusmc.nl) Genetic Laboratory Department of Internal Medicine Erasmus MC, Rotterdam Why do we study DNA variants? Population genetics: technology driven � � Biology: Time required for genotyping 1 SNP in 7.000 DNA samples from “the Rotterdam Study”: � Mechanism: understand cause of disease � 1996 6 months: RFLP, Epp tubes � 1999 3 months: RFLP, 96-well plates � Treatment: finding new potential drug targets � 2001 1 week: SBE, 384-well plates � 2003 1 day: Taqman (manual) � Prediction: � 2004 6 hrs: Taqman (automated) � (Early) diagnostics with a stable marker � 2005 3 hrs: Taqman, Deerac, “Fast” PCR � Risk of disease (vulnerability): “personalized medicine” � 2008 6 sec: Illumina 1000K array, 1000 DNAs/week � Response to treatment (medication, diet): “pharmacogenetics” 2010 0.00001 sec Illumia Hiseq, next-gen sequencing � 2015 0.000001 sec Illumina X10 1
Population genetics: technology driven Which genotyping technique to use? � Time required for genotyping 1 SNP in 7.000 DNA samples from “the Rotterdam Study”: Association study with 1 DNA variant � 1996 6 months � 1999 3 months � 2001 1 week � 2003 1 day Association study with all common � 2004 6 hrs DNA variants in one gene � 2005 3 hrs � Genome-wide association study 2008 6 sec 2010 0.00001 sec � Sequencing: causal alleles? 2015 0.000001 sec Array technology SNP Fragmented gDNA bead Address Probe 23 bp 50 bp � Illumina bead-array � Beads have probes of one SNP attached � Each bead is spotted in multifold to increase accuracy and redundancy 2
Array technology (and GWAS) based on “Linkage Disequilibrium” or “haplotypes” SNP1 SNP2 SNP3 SNP4 SNP5 SNP6 Maternal chr A C T A C T Haplotype Paternal chr G A T G C C Haplotypes movie Haplotypes and LD � Recombination is NOT random � Hot spots � In between recombination hot spots, variants are in LD � They are correlated � Often measured as r 2 � R 2 = 1 : two variants provide same information 3
How to use haplotypes: imputations Genetic architecture of traits � Correlation between variants used to ‘guess’ what the genotype of untyped variants is Hemophilia in European royalty 4
A very personal example Genetic architecture of traits Sequencing This is me De-novo mutation Genetic architecture of traits GWAS analysis Arrays Analyzing all SNPs in 1 run Visualizing results in plots Select SNPs Combine GWASs Replication Manhattan-plot Each dot represents 1 SNP Meta-Analysis of all data 5
“ERGO”: The Rotterdam Study First GWAS performed in Rotterdam Study A single-centre, prospective population-based cohort study, started 1990 LUMBAR SPINE BMD Base-line cohort = 7,983 men and women of age ≥ 55 yrs • • In 2017: 6 follow-up visits • ~1500 measurements per subject each time • Ethnically homogeneous: 99% Caucasian • Computerized GP + pharmacy monitoring 5 x 10 -8 • Study determinants and prevalence/incidence of chronic and disabling disease in the elderly: CVD, Neurodegenerative Disease, Endocrine diseases, Locomotor disease (osteoporosis, osteoarthritis), Eye • ~12.000 DNA samples available: 1990: ERGO base-line/ RSI : n=7,000 � � 2000: ERGO plus/ RSII : n=3,000 (55+) � 2004: ERGO young/ RSIII : 3,500 (45+) Rivadeneira et al., Nat Genet., 2009 EUROPE by prejudice.…….(according to USA) (From: Yanko Tsvetkov, alphadesigner.com) Most GWAS activities occur within large consortia CHARGE Rotterdam Study GEnetic Factors of OSteoporosis GENETIC INVESTIGATIONS OF ANTHROPOMETRIC TRAITS 6
LUMBAR SPINE BMD LUMBAR SPINE BMD LRP5 5 x 10 -8 5 x 10 -8 • Rotterdam Study • Rotterdam Study • ERF Study • ERF Study N=5,000 N=6,200 • Twins UK • Twins UK • deCODE Genetics • deCODE Genetics • Framingham Study Rivadeneira et al., Nat Genet., 2009 • Framingham Study Rivadeneira et al., Nat Genet., 2009 LUMBAR SPINE BMD LUMBAR SPINE BMD RANK-L C6ôrf10 OPG 1p36 LRP5 LRP5 5 x 10 -8 5 x 10 -8 MHC • Rotterdam Study • Rotterdam Study • ERF Study • ERF Study N=8,500 N=15,000 • Twins UK • Twins UK • deCODE Genetics • deCODE Genetics • Framingham Study Rivadeneira et al., Nat Genet., 2009 • Framingham Study Rivadeneira et al., Nat Genet., 2009 7
Consortia: working together does work � Much larger sample sizes can be achieved LUMBAR SPINE BMD � Go from competition to cooperation RANK-L C6ôrf10 OPG � Creates better science! 1p36 SP7 LRP5 5 x 10 -8 � But… � Only ‘cosmopolitan’ variants found � Trying to set up a call with the US, Europe and Australia is impossible � Can slow things down as you are waiting for each other • Rotterdam Study • ERF Study � Typical GWAS takes ~3-7 years N=19,125 • Twins UK • deCODE Genetics • Framingham Study Rivadeneira et al., Nat Genet., 2009 The success of consortia Example current GWAS study: age-related hearing loss � Heritability: 35-55% � Previous GWAS studies (N~3,500) have not found anything � Collected a bigger sample size (N~9,700) � Redefined the phenotype � HIGH: 4 & 8 kHz � LOW: 0.5, 1 & 2 kHz 8
Manhattan plot Chr 3 hit for HIGH phenotype ILDR1: further evidence Not always so easy… � Knock-out mice are deaf � Zebrafish knock-down have impaired auditory responses � 11 families known with autosomal recessive mutations in ILDR1 � Deafness is more pronounced at higher frequencies 9
Ethnicity should be kept in mind Example: rs776746 � Africans have more variants than Europeans/Asians � SNP in gene CYP3A5 which metabolizes clinical drugs � G allele encodes CYP3A5*3 allele � ‘Unique’ variants appeared in those that left Africa � Inactivates the gene � Adaptation to new environment � Some of these came from already existing hominids outside Africa � Frequencies of variants can differ between Ethnic groups What has/will GWAS achieve Examples of links between GWAS discoveries and drugs E D. Green et al. Nature 470 , 204-213 (2011) doi:10.1038/nature09764 10
Human Genomics facility (HuGE-F) FACILITY & RESEARCH Price GWAS measurement 800 BIOBANKING Rotterdam Study, GenR, Parelsnoer, BBMRI, many more 700 NEXT GEN SEQUENCING 600 Bench marking with top institutes of the world 28 euro : HIGH THROUGPUT ARRAYS 500 GWAS + clinical research content, GENOTYPING Collaborations in large HLA typing, pharmacogenetics 400 consortia TRANSCRIPTOMICS Functional studies in mouse models and cell 300 EPIGENETICS lines MICROBIOMICS 200 GWAS, imputation, 100 methylation analysis, exome and transcriptome BIOINFORMATICS analysis 0 www.glimdna.org 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 …..IGNORANCE CAN BE DAUNTING……EDUCATION IS IMPORTANT !! Questions Annual Courses organized by the Genetic Laboratory: in 2018: - “Genetic for Dummies” (MolMed): 15-30 participants - “SNPs and human diseases” (MolMed): 40-70 participants - “Genomics in Medicine” (ESP57; NIHES): 20-40 participants www.molmed.nl www.nihes.nl 11
Recommend
More recommend
