bcftools gtcheck output showing opposite discordance as it should

[ainsku]

So I have some bulkRNA samples I want to check are actually from the donors they are assigned to. To test this, I created a vcf file (using cellSNP-lite) where I have seven bulk samples from two donors each. The samples are from two different tissues from different time points.

When I use the -e 0 flag, and order based on discordance from lower to higher, I get the pairs from different donors as the lowest discordance and pairs from the same donor with the highest discordance. Without the -e 0 flag sorting the same way i get first some mismatches, then all of the matching pairs and at the end most of the mismatched pairs. Is there something wrong with the tool, am I running something wrong or am I not interpreting the results properly?

Here's the list of samples for both donors

donor 1
5594
A471
A472
A473
A652
C686
F948

donor 5
9179
A458
A476
A477
A654
C689
F950

merged_1_5_output.txt
merged_1_5_output_e0.txt