search

gxiaolab / scAllele

Other
28 stars 1 forks source link

No genotype information after variant calling using both single and multiple bam files. #10

Open alaminzju opened 1 year ago

alaminzju commented 1 year ago

Dear Giovanni, Thank you for your nice software. I am using scAllele for the detection and analysis of variants using scRNA-seq. I wonder how you get the information about 0/1 or 1/1 in Figure 2E of your paper. I checked your deposited results in Zenodo (https://doi.org/10.5281/ zenodo.6558451), where I found no genotype information. Moreover, I have also analyzed my data and have yet to get genotype information. I have followed all QC processes and scAllele protocols to analyze my data. I am stuck at this point. Please kindly give your explanation about it.

Advance thanks for your response.

Best Regards, Alamin Email: alamin@zju.edu.cn

giovanniquinones commented 1 year ago

The genotype information was obtained from the reference vcf file (the one we are comparing the scAllele output to) which is the GIAB HG001 vcf.

alaminzju commented 1 year ago

Can I use the GIAB HG001 vcf file for my analysis? If yes, please share the location of the vcf file.

jmfa commented 1 year ago

Hi Giovanni, I have a follow-up question regarding this matter. Could you please provide any recommendations on how to perform genotyping of single cells once variants have been identified? (or does this fall outside the scope of scAllele?)

giovanniquinones commented 1 year ago

scAllele takes a very conservative approach of not genotyping because allele-specific expression/bias can occur in the RNA-seq data resulting in wrong genotype calls. Currently, genotyping is out of the scope of scAllele but, since the allelic read counts and ratio values are given in the file, you could estimate the 'observed' genotypes.

jmfa commented 1 year ago

Thanks for the swift reply, @giovanniquinones. Right, so perhaps one idea would be to only rely on variable sites (i.e., AC!=0) - since 0/0 genotypes are trickier to infer given the allele-expression bias.

MohamedSuhail27 commented 4 months ago

Hello, I tried using the github test data (https://github.com/gxiaolab/scAllele/tree/main) to perform the variant calling using scAllele v0.0.9.4. But I didn't get the vcf header in my output vcf file.

I used the following command for variant calling: scAllele -b test_data/gm12878.chr21.bam -g test_data/hg38.chr21.fa -o scAllele_variant

Below, I have attached the output vcf for your reference. scAllele_variant.zip

Could you able to help me in this.

Thanks.