brentp / somalier

fast sample-swap and relatedness checks on BAMs/CRAMs/VCFs/GVCFs... "like damn that is one smart wine guy"
MIT License
262 stars 35 forks source link

HTML report #83

Closed Danisov closed 2 years ago

Danisov commented 2 years ago

Hi,

I ran both 10 trio, one duo and 11 unrelated samples using bam file.

Two slightly strange things I found in the html report. 1: for the duo it is well placed among parent-child but the point color corresponds to the color as unrelated samples. 2: A female sample, it is well palced among the female samples but the point coloring (gray) is different compared to the others female.

Do you have an idea ??

brentp commented 2 years ago

1: this should not be the case, but you can try adding a fake dad to the kid. e.g. "fakedad". and then add a sample line for the fakedad. e.g.: $fam_id fakedad -9 -9 1 -9. Can you show the lines for the 2 samples?

2: likewise, can you show the line in the pedigree file for this sample?

Danisov commented 2 years ago

Hi @brentp Please find the first line corresponds to the duo (samplaA and sampleB). The second corresponds to the sample long with original sex pedigree unknown (sample1)

family_id   sample_id   paternal_id maternal_id sex original_pedigree_sex   gt_depth_mean   gt_depth_sd depth_mean  depth_sd    ab_mean ab_std  n_hom_ref   n_het   n_hom_alt   n_unknown   p_middling_ab   X_depth_mean    X_n X_hom_ref   X_het   X_hom_alt   Y_depth_mean    Y_n
sampleA sampleB         1   male    39.6    7.6 39.6    7.6 0.53    0.39    4594    6869    5613    308 0.009   19.78   356 181 0   175 20.33   15
sample1 sample1             unknown 35.6    7.3 35.6    7.4 0.53    0.39    4060    6668    5060    1596    0.071   35.41   341 105 138 98  0.00    0
brentp commented 2 years ago

This line: sampleA sampleB 1 male is a single sample. Where's the other sample in that family? It will need to have the same family id (sampleA) for somalier to match it.

This line: sample1 sample1 unknown shows that there is no sex specified in the column of the pedigree file.

So you just need to update the pedigree file that you send to somalier to include the sex and the include some value for sampleA.

Danisov commented 2 years ago

Hi @brentp sorry for the delay of my answe I used pedigree file obtained by somalier. For a reason of validation of the manipulations within our laboratory, was asked to determine the sex and the relatedness within these samples without having an idea on their sex and real relatedness that's why I used the pedigree file obtained by somalier.

The question I ask myself: is that for the sex and the relatedness was well determined in the pedigree file obtained by somalier with the exception of the samples that I had mentioned

brentp commented 2 years ago

Hi, I'm not able to understand your response. If you can respond to my previous message point-by-point, perhaps I can help. Or if I miss a question in your latest, I will attempt to help if you can clarify.