The SNP number of output file dose not match the stdout

[xinyu-c9]

I was hoping I could ask for your help with something. Recently, I ran the DBSLMM algorithm and found that the SNP number of output file does not match the stdout.

The file sizes of the result file are listed below, from which you can see the file size of chr10, chr13, and so on are significantly lower than other files.

Following are the result files of chr10 and chr13

Following are the stdout of chr10 and chr13

according to the stdout file, the output SNP site number for chr10 and chr13 should be 37139 and 29556, but the result file only contains a few SNPs.

Then after counting the number of lines in result files, and comparing it with stdout(I add up the large effect number and small effect number), I found the difference in chr3 5 7 10 11 13 15 16 17 18 20. Some of them are very different, those with small file sizes in Figure 1 belong to this type. The other part is only a few hundred gaps. This is so weird.

This is my code when ran the DBSLMM algorithm:

export OMP_NUM_THREADS=1
export LD_LIBRARY_PATH=/software/conda/envs/wdl/lib:$LD_LIBRARY_PATH
for i in <all 22 chromosome gwas file name>
do
        j=`basename $i | sed -e 's/summary_chr\(.*\).assoc.txt/\1/'`
        /software/conda/envs/wdl/bin/Rscript \
        /software/DBSLMM/DBSLMM/software/DBSLMM.R \
            --summary $i \
            --outPath ./ \
            --plink /software/plink/plink \
            --model DBSLMM \
            --dbslmm /software/DBSLMM/DBSLMM/scr/dbslmm \
            --ref /LD_reference/DBSLMM/plink  \
            --n $N \
            --nsnp $n_snp \
            --type t \
            --block /software/DBSLMM/DBSLMM/block_data/eur/chr$j.bed \
            --h2 0.012581797  \
            --h2f 1 
done

if you had any tips or suggestions for how to approach the issue. I would appreciate any help you could offer. Thank you so much in advance!

[biostat0903]

Hi Xingyu,
Thanks for your consideration for DBSLMM. I am also weird for the error.
I recommend you can solve your problems in two ways:

1. check the block file of the wrong chromosomes;
2. use the online version of DBSLMM: https://www.pgs-server.com/

Best,
Sheng