Combining Multiple Genetic Estimates of N_e.

Area of Science:

• Population Genetics
• Quantitative Genetics
• Bioinformatics

Background:

• Estimating contemporary effective population size (N_e) is crucial in population genetics.
• Multiple genetic methods exist, but combining their estimates can increase precision.
• Current methods for combining N_e estimates have limitations.

Purpose of the Study:

• To develop an optimal inverse-variance weighting scheme for combining N_e estimates.
• To address the limitations of previous weighting approaches based on var( ${\hat{N}}_e$ ).
• To introduce a new method using weights inversely proportional to var(1/ ${\hat{N}}_e$ ).

Main Methods:

• Developed a novel weighting scheme based on the inverse variance of 1/N_e.
• Applied analytical and numerical methods to evaluate the weighting scheme.
• Introduced new software, ComboNe, for calculating combined N_e estimates.

Main Results:

• The new weighting scheme, var(1/ ${\hat{N}}_e$ ), is robust and suitable for combining N_e estimates.
• Existing theory supports robust estimation of var(1/ ${\hat{N}}_e$ ) for temporal and LD methods.
• LD and sibship methods show varying correlation depending on dataset size, impacting optimal combination.

Conclusions:

• The new inverse-variance weighting approach improves the precision of effective population size estimates.
• The ComboNe software facilitates the optimal combination of N_e estimates from different genetic methods.
• This work provides a robust framework for integrating diverse N_e data.