impMKT

impMKT applies imputation within the McDonald and Kreitman framework to improve detection and quantification of recurrent positive selection from DNA sequence data at the gene level.

Key Features:

• McDonald and Kreitman framework: Extends the McDonald and Kreitman test for gene-level analyses of DNA sequence data.
• Imputation of low-frequency variants: Uses imputation techniques to correct the underestimation of positive selection caused by low-frequency slightly deleterious variants.
• Increased detection power: Demonstrates higher statistical evidence of positive selection, with reported detection increases of ~50% in Drosophila and ~157% in humans in comparative studies.
• Estimation of adaptive substitutions (α): Produces more reliable estimates of the proportion of adaptive substitutions (α) using fewer genes than conventional MK implementations.
• Validation datasets: Evaluated on both simulated and empirical datasets from human and Drosophila genomes.

Scientific Applications:

• Gene-level detection of positive selection: Detects and quantifies recurrent positive selection in protein-coding genes from DNA sequence data.
• Estimation of adaptive substitution rates: Estimates the proportion of adaptive substitutions (α) for comparative and population-genomics studies.
• Comparative genomics: Applies to empirical comparisons between species, exemplified by human and Drosophila genome analyses.
• Method validation and simulation studies: Supports evaluation of selection inference methods using simulated datasets.
• Evolutionary hypothesis testing: Enables testing of gene-level evolutionary hypotheses about adaptive evolution.

Methodology:

Implements imputation within the McDonald and Kreitman test to adjust for low-frequency slightly deleterious variants and was validated on simulated and empirical human and Drosophila genome datasets.