Individuals genetically similar

Question

[Shuki]

Hello to all, I have just started in the field and I would like to have please some suggestions on the analysis that I wish to make!

I have a table of SNPs from genotyping with in row the individuals and in column the SNPs. The table contains values from 0 to 3 (0 missing data, 1 reference allele, 2 alternative allele, 3 heterozygote). The file is in txt format. I want to identify on the basis of this genetic information the similar or more genetically related individuals. Does anyone have an idea on the steps to follow?

Thanks

Answer 1

[patrickdm]

Hello, you could compute a matrix of pairwise Hamming distances between your individuals' alleles calls (i.e. see https://datagy.io/python-hamming-distance) and cluster them (with PHYLIP neighbor or another tree reconstruction tool working from distance matrices (MEGA does that too, but its input file format is somewhat unfriendly).

If you have many missing data calls, and you want to correct the pairwise distances to account for them, you could look at how it is implemented in: Galpern P, Manseau, M, Hettinga P, Smith K, and Wilson P. (2012) allelematch: an R package for identifying unique multilocus genotypes where genotype error and missing data may be present. Molecular Ecology Resources 12:771-778

Hth.