How Can I Estimate Effective Population Size From Linkage Disequlibirum (R^2) Result?

Question

[mary]

Dear all

I have a genotype data for SNP in cattle, I calculated r^2 by plink and now I want to calculated effective population size of my population. the explanation for how calculate Ne by LD result are not clear in literature, I thinks that I have to use this formula: Ne = (1/4c)*([1/r^2] - 1 ) , The age of Ne for any distance is estimated by 1/2c. is there any one that can offer me some paper or explain clearly how i should calculated Ne from this formula. I know that I can calculate recombination rate by phase or assume that 1cM=1Mbp, and also I know that I should use r2 but I don't know if I calculated r2 in bin distance how can I chose r2 for each cM?

thanks a lot

[GabrielMontenegro]

Any luck using LDNe? I have the same problem myself, I am using genome-wide SNP data and I am not sure if that program can handle such big datasets.

Answer 1

[AGS]

Waples (1991) used N_e = 1/3(r^2 - (1/s)) where s is the harmonic mean of the sample size. England et al (2006) showed that this estimator is biased for sample sizes under ~ 100.

1. Waples RS (1991) Genetic methods for estimating effective population size of Cetacean populations. In Genetic Ecology of Whales and Dolphins (ed. Hoelzel AR), pp. 279–300. International Whaling Commission, Special Issue 13, Cambridge.
2. Estimating effective population size from linkage disequilibrium: severe bias in small samples Phillip R. England, Jean-Marie Cornuet, Pierre Berthier, David A. Tallmon and Gordon Luikart. Conservation Genetics, 2006, Volume 7, Number 2, Pages 303-308

Answer 2

[Josh Herr]

If you're estimating effective population size I would recommend reading this classic paper. I read this newer paper a couple months ago. I don't know the insides and out of the analysis, but I would recommend the LDNe program (citation) written in visual basic.