Bar Plot from Scientific Research

Results of the GWAS analysis. In each dataset, 5%, 10%, and 20% of the strains were mislabeled and the GWAS analysis was repeated. The measure of success ( x -axis) was the mean of ranks of the true causal genes. All genes were first ordered by their p -values (or F1 values or standardized residuals) and then the average rank was computed for the causal genes. Four existing mGWAS tools and aurora were used for these analyses.

DOI