Power of neutrality tests for detecting natural selection

Area of Science:

• Population genetics
• Evolutionary biology
• Genomics

Background:

• Detecting natural selection is a key area in population genetics.
• Numerous tests exist for identifying natural selection using genomic data.
• The effectiveness of these tests is influenced by evolutionary factors like selection timing, strength, allele frequency, and demographic history.

Purpose of the Study:

• To quantitatively assess the power of four common natural selection detection tests: Tajima's D, Fay and Wu's H, relative extended haplotype homozygosity (rEHH), and integrated haplotype score (iHS).
• To investigate the relationship between evolutionary parameters, demographic models, and the performance of these tests.
• To expand the understanding of approaches for detecting natural selection.

Main Methods:

• Simulated genomic data under various evolutionary parameters and demographic models.
• Evaluated the statistical power of Tajima's D, Fay and Wu's H, rEHH, and iHS tests.
• Compared test performance across a range of selection strengths, timings, allele frequencies, and demographic scenarios.

Main Results:

• Each of the four tested methods demonstrated effectiveness only within specific parameter ranges.
• Significant ranges of evolutionary parameters were identified where none of the tested methods effectively detected selection.
• The parameter spaces yielding the highest power for each test generally aligned with previous empirical findings.

Conclusions:

• Current genomic tests for natural selection have limitations and do not cover all possible evolutionary scenarios.
• Our understanding of adaptation is potentially restricted by the limited scope of existing detection methods.
• Further research is needed to develop more robust methods for detecting natural selection across diverse evolutionary contexts.