Simulation theory Not even counting recombinants AAbb aaBB AaBb With modest marker spacing in a human study LOD of 3 is 9 likely to be a false positive inter mate Two loci Need one dominant allele at each locus to get phenotype AABb AaBb aaBb AaBB aaBB Flower color Genotype at marker close to A locus Aabb NOT complete coinheritance purple white Top allele 3 1 Bottom allele 2 3 Affected sib pair method 2 2 2 3 4 4 1 3 2 2 2 2 1 4 1 4 Sib pairs Observed Expected under null Same allele Different allele 2 1 2 2 0 1 2 2 Unlike cystic fibrosis and Huntington s disease most traits are not yes or no Doesn t require you to know dominant or recessive doesn t require large pedigrees Model free a good thing E g blood pressure www jax org staff churchill labsite pubs qtl pdf 1 Distributions Distributions Distributions Environment and error 2 What if Plain water What if Salt water Exact same mouse every day for 6 mo What if Exact same mouse every day for 6 mo What if Many clones identical twins Time of day Change in cage mates Age Reproductive cycle 3 What if Imagine a cross AA x aa Experimental error random variation Aa F1 Aa x Aa Many clones identical twins Time of day Change in cage mates Age Reproductive cycle AA Aa aA aa F2 Single locus dominant Single locus dominant F2 s F2 s 75 25 4 Single locus dominant Single locus dominant Aa aA AA F2 s Aa aA AA aa 75 25 F2 s aa 75 25 A given genotype does not guarantee exactly the same phenotype every time More generally 1 locus AA x BB AB F1 AB x AB AA AB BA BB F2 5 1 locus 1 locus AB BA AA 25 50 25 1 locus incomplete dominance 25 25 50 50 25 1 locus incomplete dominance AB BA AA BB AB BA BB AA 25 25 BB 50 25 Remember Fig 3 2 6 Another representation 1 locus incomplete dominance AB BA AA AB BA BB AA BB Effect of having a B 25 50 25 AA AB BA BB 1 locus incomplete dominance 25 25 AA AB BA BB AB BA BB 50 25 1 locus incomplete dominance AB BA AA 50 25 Effect of having a B AA AA AB BA BB 25 BB 50 25 Effect of having a B AA AB BA BB Effect of a B allele is the same regardless of genotype additive 7 1 locus complete dominance AB BA AA 1 locus complete dominance AB BA AA BB 75 25 BB 75 25 1 locus complete dominance All these are examples of a single locus controlling variation in a quantitative trait AB BA AA BB 75 25 But usually more complicated Dominance is a kind of epistasis nonadditive 8 A real example A real example CC x SS CS F2 s F2 s CS x CS CC SS CS SC F2 s A real example F2 s A real example F2 s C3H parent C3H parent SWR parent 9 Not 1 locus dominant or 1 locus incomplete dominance or F2 s Doesn t look like this C3H parent Test for linkage to markers to find unknown genetic determinants as before SWR parent Quantitative trait linkage test Test for linkage to markers to find unknown genetic determinants as before Mechanics of test are different F2 s Inheritance at a marker 10 Quantitative trait linkage test Quantitative trait linkage test Mean BP across all F2 s that carry two SWR alleles at marker F2 s F2 s Alleles are named after parent Quantitative trait linkage test F2 s Mean BP across all F2 s that carry two C3H alleles at marker Quantitative trait linkage test F2 s Not counting recombinants Statistical test for goodness of fit 11 Locus effect vs parents Locus effect vs parents Locus effect F2 s split out by genotype C3H parent F2 s F2 s C C at C S at marker marker F2 s SWR S S at parent marker C3H parent F2 s F2 s C C at C S at marker marker F2 s SWR S S at parent marker Alleles are named after parent Locus effect vs parents Locus effect vs parents Homozygotes do not look like parent Each parent strain C3H parent F2 s F2 s C C at C S at marker marker F2 s SWR S S at parent marker C3H parent F2 s F2 s C C at C S at marker marker F2 s SWR S S at parent marker 12 Locus effect vs parents Locus effect vs parents Homozygotes do not look like parent What do you infer C3H parent F2 s F2 s C C at C S at marker marker F2 s SWR S S at parent marker Homozygotes do not look like parent What do you infer C3H parent F2 s F2 s C C at C S at marker marker F2 s SWR S S at parent marker A single varying locus does not explain the data 1 locus controlling trait What if One mouse family Magnitude of error bars spread within group has not changed Locus effect is weaker C3H parent F2 s F2 s C C at C S at marker marker F2 s SWR S S at parent marker 13 Correct interpretation Correct interpretation Effect of having an S allele C3H parent F2 s F2 s C C at C S at marker marker F2 s SWR S S at parent marker Difference between S and C at this locus has a causal role in blood pressure variation but effect is modest Correct interpretation C3H parent F2 s F2 s C C at C S at marker marker F2 s SWR S S at parent marker Difference between S and C at this locus has a causal role in blood pressure variation but effect is modest Complex traits one family mouse model Effect of having an S allele Most loci underlying human disease look like this C3H parent F2 s F2 s C C at C S at marker marker F2 s SWR S S at parent marker 14 Complex traits Complex traits Genetic differences at both loci affect the trait one family mouse model Complex traits one family mouse model Complex traits Each locus responsible for half Each locus responsible for half Depends on the model one family mouse model one family mouse model 15 Complex traits one family mouse model Complex traits one family mouse model Each locus responsible for a third Complex traits If 5 loci each responsible for a fifth 10 loci The more loci the smaller the effects and the harder to detect Complex traits One common result of a linkage study is no significant linkage anywhere Genetic complexity is the rule simple 1or 2 locus models are the exception 16 We haven t talked about humans lately New York Times Nov 11 2007 With model organisms can always study a single cross family with lots of progeny so better statistical power to detect weak loci …
View Full Document
Unlocking...