See caravagnalab/ProCESS#149
The pre-sampling approach can be avoided by splitting index buckets in multiple sequential files. The order of the loci in each bucket and in files should be random. The probability of a loci will depend on its position on the bucket as it is for the pre-sampling technique.
See caravagnalab/ProCESS#149
The pre-sampling approach can be avoided by splitting index buckets in multiple sequential files. The order of the loci in each bucket and in files should be random. The probability of a loci will depend on its position on the bucket as it is for the pre-sampling technique.