Title: Part 2: "Hitchhiking mapping to find quantitative trait loci (QTL) associated with phenotype differential of resource populations... a work in progress..."
Speaker: Bill Muir; Department of Animal Science, Purdue University and Visiting Professor Department of Medicine IUPUI, Indianapolis, IN

Place: LILLY Hall G126
Date: October 20, 2015; Tuesday
Time: 4:30pm

(Part 1 of this talk took place on September 22, 2015; same title and abstract)
With the advent of economical, high throughput, whole genome NGS technologies, new opportunities are now available to identify trait-associated sequence variations with high efficiency and high resolution that were not (previously) detected by QTL methods. Using NGS technology, we can directly examine detailed patterns of variation in the genomic DNA for evidence of selection by comparing the sequences of divergent selected lines. Selection for a trait brings about phenotypic changes through changes in allele frequencies (polymorphisms) associated with genes that cause variation in that trait. Polymorphisms can be single nucleotides (SNPs), copy number variations, or genomic rearrangements. Alleles that contribute to the selected trait increase in frequency, while those that reduce it are selected against; eventually, key contributing polymorphisms can become fixed in the selected populations. However, selection on the beneficial variant affects more than just that single polymorphism; other surrounding variations close to the selected ones will be co-selected, and are said to be "hitchhiking" (Maynard-Smith & Haigh 1974). Because of hitchhiking, as selection drives beneficial alleles to fixation, surrounding variations are also driven to fixation, leaving large areas of homozygosity known as "signatures of selection", and alternatively as a "selective sweeps". Thus, specific patterns of selection acting on individual loci are detectable through hitchhiking mapping (Schlotterer 2003).

There are a number of statistical problems associated with using this method, including confounding due to similar effects of selection and random genetic drift on genomic patterns, experimetnal design (levels of replications), which metric to use, and hypothesis testing (Nielsen, 2005; Walsh, 2008). I will be discuss these issues in relation to trying to find QTL associated with the human disease of alcoholism via animal models (rats) that are long-term selected (60+ generations) for, and against, alcohol drinking behavior. We recently sequenced indivudals from these lines as part of the Indiana Alcohol Research Center (IUPUI) NIH funded center grant; data analysis is in progress. The goal of this talk is to stimulate discussion and ideas about the challenges of these data, and to gain potential statistical solutions to this complex problem.

Associated reading:
1. Nielsen, R. 2005. Molecular signatures of natural selection. Annu. Rev. Genet. 39:197-218

2. Smith, J.M.and J.Haigh. 1974. Hitch-Hiking Effect of A Favorable Gene. Genetical Research 23:23-35

3. Walsh,B. 2008. Using molecular markers for detecting domestication, improvement, and adaptation genes. Euphytica 161:1-17

Click here www.stat.purdue.edu/~doerge/BIOINFORM.D/FALL15/sem.html for a full schedule of BIOINFORMATICS SEMINARS, past and present.