Computational protocol: QTL Analysis of Dietary Obesity in C57BL/6byj X 129P3/J F2 Mice: Diet- and Sex-Dependent Effects

Similar protocols

Protocol publication

[…] Linkage analysis was conducted using the algorithms implemented in R/qtl version 1.08–56 , in four successive steps: (a) a genetic map was estimated from the marker genotypes from the F2 mice, (b) covariates were identified using regression methods, (c) main effect QTLs and pairwise interactions were identified, and (d) the covariates and all main effect and interacting QTLs were incorporated into a model to determine how much phenotypic variance could be accounted for. Descriptive statistics and the initial assessment of covariate effects were conducted using Statistica 8.0 (StatSoft, Tulsa, OK). [...] The goal of the genotype association mapping analysis was to find regions (within the locus boundaries) that are the most likely to contain the genetic variants that account for the observed QTL. Therefore, we extracted percent body fat phenotype data from the Mouse Phenome Database from two studies: a 40-strain survey of 14- to 18-week-old mice fed a standard laboratory chow (low-energy diet; ) using data originally collected in our laboratory , and a 43-strain survey of 15– to 17-week-old mice fed a high-energy diet for 8 weeks conducted at the Jackson Laboratory . Percent body fat was measured in both strain surveys using the same method employed in our linkage analysis (Experiment 1).We examined these data in conjunction with strain-specific nucleotide variants as described below. Data analysis was conducted using two statistical approaches as implemented in the program GEMMA , following the guidelines suggested by their developers and with imputation of 4 million SNPs genome-wide. Sex was used as a covariate in the analysis. Results were filtered by the locus boundaries, and the top 5% of nominally significant results (P<0.05) were extracted from each of the four QTL regions.We also evaluated the genotype association mapping results in conjunction with the pattern of results from similar studies (i.e., multiple cross mapping), paying particular attention to the possibility that the causal variant might have arisen recently and thus would not be detected in other QTL mapping studies. Therefore, we extracted all previously published matching QTLs for comparison from the Mouse Genome Informatics database. […]

Pipeline specifications

Software tools R/qtl, Statistica, GEMMA
Databases MGI MPD
Applications Miscellaneous, WGS analysis, GWAS
Organisms Mus musculus, Homo sapiens