DNA Unit 2 Lecture 1

DNA Unit 2 Lecture 1 Chromosomes: recombination during meiosis, paternal and maternal lines up with physical linkage that has a mixture in the next chromosome ● Genes/alleles close together in chromosomes, tend to stay together in offspring because the chance of crossing over decreases in a smaller space Linkage Disequilibrium (LD): a statistical correlation between alleles at different genes ● ● ● ● NONRANDOM results from genes located close together in the genome Every time there is a A, there is B, when a, there is b, (perfect association) close together at the chromosome so recombination doesn’t break them up biased association of a marker, helps identify ancient mutations or specific alleles ● Recombination breaks linkage disequilibrium ○ Ex, 1 ancestral haplotype that gains a mutation, now 2 haplotypes exist AB to AB and Ab Linkage Equilibrium: lots of recombination at the A and B loci ● ● ● ● RANDOM association between a marker and a disease A can B or b, a can be B or b, (frequency is by chance) separate in population If know A loci, don’t know b loci (all mixed up) Chromosomes further apart, more recombination breaks up associations Genome-Wide Association (GWAS): large enough sample, certain SNPs might be associated with a phenotype, giving clues as to where in the genome the genes causing the phenotype may reside. Correlation between SNPs and a phenotype. Map location of SNPs in genome to map to see what is causing the phenotype ● SNP 3: high frequency with disease and low without ● Continuous trait of disease liability where control is unaffected and affected is case ○ What is frequency of particular allele? ● Odd Ratios: measures the association of a genotype and risk of developing a disease. When there is more knowledge of what gene affects the disease, this increases chance of winning/probability of finding the affecting gene ○ Ratio between the odds of individuals having a phenotype associated with a specific allele and the odds of the same phenotype for individuals who don’t have that same allele ● SNP for cancer: who has certain C or T at location for odds of getting cancer ○ Having a C SNP has a higher chance/odd ratio of getting cancer than T SNP ○ Y axis is Odds ratio, X axis is location, region of the SNPs that stand out that correspond to the phenotype: all make up the map ○ The SNPs that are statistically associated with the phenotype = GWAS hits ○ Missing heritability is likely due to our inability to detect rare alleles/genes of small effect in GWAS studies, whereas the large effect is easier to detect ■ Due to genes/diseases being less common and each GWAS detects different things and misses a lot of stuff ■ Heritability is higher in twin studies than in GWAS = more accurate ● Used to compare the heritably of traits in identical vs. Fraternal traits to see the proportion. Traits with high heritability are likely to yield significant findings in GWAS. Used to see how genetic factors interact with environmental influences on a trait. Learns more about complex traits ○ Currently, GWAS is getting larger/rapidly and gap of missing genes is smaller ○ Higher sample size doesn’t level off number of GWAS hits (still early phase) ■ GWAS hits are like the stars, the harder you look the more you find ○ Heritability and genetics are the basis for everything and shows correlation of genotype and phenotype ○ Threshold low for Manhattan plot of generalized risk tolerance: a lot of people are risk-averse (less likely to take risks), is a common trait, is a social construct and has a large sample size ■ Not studying the trait/understanding bio of humans (social construct) ○ The higher on graph, the more likely it is that gene for clusters of associations for gene of interest (markers) with higher statistical significance. As sample size increases, more statistical significance results for more genes (peaks)