Single Cell genotypes
qian0001 opened this issue · comments
Hi,
I am using your SCAN-SNV to find the somatic variants from our WGS. We are concerned about the somatic genotype call results for the single cells. I am attaching 4 example record that passes somatic$pass (i.e., somatic$pass == TRUE) in table somatic from one of somatic_genotypes.rda files. I have transposed row/columns for better displays.
The 4 examples show the SingleCell genotypes (row 11) as 1/1, and coverage depths (ref.1,alt.1) as (0,10), (1,14), (0,27) and (0,15).
I understand that these results are coming from SCAN-SNV’s first step, GATK, for selecting candidates variants in your pipeline. The genotype calls and coverages may not be used in later steps. Should I assume these four cases are all true somatic mutations with heterozygous genotypes even though the raw data points to homozygous alternative alleles?
I also like to ask why final genotype call results would contradict the initial GATK genotype calls. Those coverages strongly suggest that the somatic variants are homozygous, as almost all the reads have the “alt.1” alleles. Put it in another way, is it true that the AB model will be so strong/confident to call it a heterozygote when there are 0 ref alleles and 27 alternative alleles?
This is important as we are looking at somatic signatures and need the accurate genotypes for the single cells. Please advise.
Thanks you very much.
Yong