multiple peaks (bimodal) in Individual Call Rate distribution

[Gabriella Scatà]

Hi everyone,

I am writing just to know whether any of you ever got a bimodal distribution in the individual callrate plot (obtained with gl.report.callrate).

If I filter the dataset with a Locus CR >0.50 (~17% loci removed), I lose the first peak in the Individual CR bimodal distribution, but I wonder whether this missingness level is due simply to poor quality DNA or to actual genetic differences as 50% of individual are missing ~ 17% loci. So, I am just afraid I am removing informative loci in this way...

Does any of you have any suggestion on how to answer this question?

I would appreciate any suggestion!

Thanks

Best

Gabriella