Lecture 21.docx

5 Pages
Unlock Document

University of Toronto St. George
Ecology & Evolutionary Biology
Stephen Wright

Lecture 21:  The genetic basis of phenotypic variation: the scale of the problem:  Human genome: 3.2 billion  Any 2 individuals are on average 99.9% identical (3.2 million differences)  The differences account for heritable phenotypic variation – including disease susceptibility variation  30, 000+ genes  Estimated 11 million common (>1%) SNPs plus many many rare variants  Which SNP affects which disease?  Heritability:  Variation of Phenotype = Variation (Genetic factors) + Variation (Environmental factors)  If variation in the phenotype is at least partly due to genetic variation, then there is heritability for that trait  Heritability: the genetically determined proportion of trait variation in the population, the percentage of variation in a trait that is due to genetic factors  The presence of heritability for the trait is a prerequisite for identifying genetic variants underlying trait variation  Mendelian disease:  rare disorders  one single gene influences the phenotype very little environmental factors  Discrete phenotypes caused by alleles segregating at a single genetic locus  Highly heritable – allelic variation at the single locus is sufficient to cause disease  Phenotype is an accurate predictor of genotype  Polygenic disease: common disorders  Variation in phenotype in the population is caused by alleles segregating at multiple genetic loci  The more loci affecting the trait, the greater the number of possible multi-locus genotypes. Different combinations of loci affect the same phenotype  E.g. for n loci with 2 alleles at each locus, there are 3 possible genotypes at each locus thus 3^n possible multi-locus genotypes  Polygenic traits: the effects at different loci add together to result in the phenotype, the more loci affecting the trait the more genotypes there are with the same phenotype  More and more number of alleles = more possible genotypes = more possible quantitative risk, 100 genes affecting heart disease = a lot of risk strategies  Mapping and Identifying disease-causing variation:  Family history is a strong risk factor for nearly all diseases, suggesting that inherited genetic variation matters (i.e. there is a non-zero heritability) for genesis of disease  The goal of genetic mapping is to establish that variation at a genomic locus is correlated with the trait phenotype  The hope is that identification of the genes and alleles will 1) predict disease susceptibility (do you have the disease allele) and 2) lead to treatment options through a better understanding of the disease biology (molecular pathways that lead to the disease)  Disease Mapping Methods: 1. Linkage mapping: marker trait correlation assessed in pedigrees/families, Mendelian 2. Genome wide association mapping: marker trait correlation assessed in the population. What markers are correlated with disease risks?  Linkage approaches are successful for Mendelian traits, despite the limited recombination events in the pedigree  Linkage-mapping and polygenic traits have mixed results  No clear genotype-phenotype relationship.  The genetic basis of the phenotype is heterogenous and our ability to look at large amounts of progeny is low: A bunch of alleles that are increasing and decreasing risk of heart disease. Linkage studies typically find large genomic regions that significantly affect the trait, but the results are usually not replicated.  Different genomic regions are mapped in different samples of pedigrees  GWAS:  Correlation between genetic variant and trait on a population level  E.g. human height. Are the individuals with g allele taller than individuals with t allele? Kind of like a t-test. The blue dots indicate regions with a significant effect  Problems: (3)  False negative: many biological associations do not reach stringent significance threshold  False positive: The more tests you do the more false positive you expect to see just by chance. At significance level = 10^-5 (p=0.05), the number of false positives with 500,000 tests is 5 and 10 for 1,000,000 tests  Suppose in your experimental design (1000 cases and 1000 controls, genotypes at 500,000 SNPs) you are able to detect 10 true associations at significance level = 10^-5. Thus only 50% of the associations are real.  Assumes that common diseases are caused by common variants, not going to work if everyone has the disease for different reason and limits the power to detect rare effects  Common disease common variant hypothesis  Rare variants of small effect are very hard to identify by genetic means. Few examples of high effect common variants influencing common diseases  Only looks at: rare alleles causing M diseases at high Ne, low frequency variant with intermediate effect at interm
More Less

Related notes for EHJ352H1

Log In


Don't have an account?

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.