I'm struggling to understand the Watterson estimator. Can someone explain me how do I compute Watterson's theta across the whole genome of each population in a large SNP dataset? If I have two populations with the same number of SNPs and the same number of individuals, will the Watterson's theta be the same too? Thank you in advance.
For stuff like this I like to use the software ANGSD (http://www.popgen.dk/angsd/index.php/Thetas,Tajima,Neutrality_tests). Since Watterson's is a measure of genetic diversity, having the same number of SNPs and individuals between your populations should produce similar estimates.
Thanks @Justin ! Your answer was helpful. Would it be possible to compute Watterson estimator on a beagle file format? I don't have easy access to BAM files (required for ANGSD). All I have is a table in which each row is a different SNP (up to 6 million rows) and every two columns I have the two alleles for a different individual (i.e., biallelic dataset). My aim was to manually calculate the Watterson estimator using its formula (https://en.wikipedia.org/wiki/Watterson_estimator). But I can't understand the logic behind that formula. Thank you again!
Hi Santiago
The Watterson estimate should be used with non-recombining sequences (stretches of DNA), not with SNP data (see, for example, Arlequin manual, page 120)
Pablo
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Bioinformatics