On Fisher's exact test for detecting gametic disequilibrium between DNA polymorphisms

Area of Science:

• Population genetics
• Statistical genetics
• Molecular evolution

Background:

• Fisher's exact test is widely used for detecting nonrandom associations between DNA polymorphisms.
• This test is known to be conservative, potentially leading to underestimation of true associations.
• Tocher's correction offers a theoretical improvement but is seldom applied in practice.

Purpose of the Study:

• To compare the practical performance of Fisher's exact test with and without Tocher's correction for detecting gametic associations.
• To evaluate the impact of using Tocher's correction on identifying significant associations in population data.
• To compare these methods with the standard chi-square test for independence.

Main Methods:

• Analysis of 1566 pairs of restriction fragment length polymorphisms (RFLPs) from human and Drosophila datasets.
• Application of Fisher's exact test with and without Tocher's correction.
• Comparison of results with the standard chi-square test for independence in 2x2 tables.

Main Results:

• Significant differences were observed between the tests in detecting gametic associations.
• The exact test with Tocher's correction identified 2-3 times more significant associations in some gene regions compared to the uncorrected test.
• The standard chi-square test showed comparable detection rates to Tocher's corrected test.

Conclusions:

• The widely used Fisher's exact test may seriously underestimate genetic disequilibrium in populations.
• Tocher's correction is a more sensitive method for detecting gametic associations.
• The standard chi-square test offers similar sensitivity to the corrected exact test and may be a practical alternative.