Plink 1.9 file-LD based SNP pruning- -problems with autoconvert option??

[Hsarin]

Hi, I'm trying to perform LD based SNP pruning on my data so I could use the output SNP list on determinating the PCA's from the dataset. The dataset is imputated genotype data.The command I've been trying to use is 

/gpfs/homes/hsarin/PLINK_1.9/plink1.9 --bgen ../dilgom_1000g_chr1.bgen –sample ../dilgom_1000g_chr1.sample --indep-pairwise 100 20 0.1 –out SNPs_pruned

In result, I however get temporary .bed .bim and .fam files(and also the .log file) but not the result files : prune.in or prune.out. Could somebody help with this what might be the problem? Really appreciate the answers!

[Hsarin]

The log file says: Total genotyping rate is 0.982489.

489301 variants and 2660 people pass filters and QC.

Note: No phenotypes present.

--indep-pairwise: Ignoring 489301 chromosome 0 variants.

Error: Too few valid variants for --indep-pairwise.

[Christopher Chang]

--indep-pairwise requires chromosome codes and bp positions to function, since it uses a sliding window.  The "chromosome 0 variants" line in the log implies that chromosome codes are missing from the .bgen.

If they're actually present, can you send me the .bgen and .sample files so I can investigate further?  Thanks!

[Hsarin]

Ok, so you were right. The chromosome codes are actually missing and also the two first columns are NA and ----- before rsXXXXX. What would be the easiest way to change file to the right form? So I have to delete the first to columns and then add the chromosome codes? Really appreaciate your help!

[Daniel Poveda Martínez]

did you fix this? I also have the same problem

--indep: Ignoring 92460 chromosome 0 variants.

Error: Too few valid variants for --indep.