Estimating IBD using MLE

[BELKHIR]

Hi,

I'm trying to run snpgdsIBDMLE function from the SNPRelate package on a seqArray obtained from your example :
file <- seqOpen(seqExampleFileName("KG_Phase1"))
mlibd = snpgdsIBDMLE(file,num.thread=8)

This return :
Identity-By-Descent analysis (MLE) on genotypes:
Calculating allele counts/frequencies ...
[==================================================] 100%, completed, 0s (process 1)
Excluding 122 SNVs (monomorphic: TRUE, MAF: NaN, missing rate: NaN)
Working space: 1,092 samples, 19,651 SNVs
using 8 (CPU) cores
Error in snpgdsIBDMLE(file, num.thread = 8) : Invalid position in CIndex.

However if I convert the seqArray file object to SNP GDS file this function works fine.

I'm missing some thing ?

Regards,

[zhengxwen]

If you specify a list of SNP IDs in snpgdsIBDMLE(), it works.

[BELKHIR]

Thank's !
It can be a temporary solution.

Best regards

[BELKHIR]

Hi,

There is still a problem if the function is called for a subset of snp:

file <- seqOpen(seqExampleFileName("KG_Phase1"))
variant.id <- seqGetData(file, "variant.id")
mlibd = snpgdsIBDMLE(file, snp.id = sample(variant.id, 100), num.thread=8)

Identity-By-Descent analysis (MLE) on genotypes:
Calculating allele counts/frequencies ...
[==================================================] 100%, completed, 0s (process 1)
Working space: 1,092 samples, 100 SNVs
using 8 (CPU) cores
Error in snpgdsIBDMLE(file, snp.id = sample(variant.id, 100), num.thread = 8) :
Invalid position in CIndex.

Best regards,