I am sequencing 31 libraries each prepared from 1/6th of the DNA in a single cell from the same cow. Therefore ~95% of positions in each library are expected to be homozygous and this is the case. I also have sequence of the same animal from bulk genomic DNA and I am analysing the data only at positions that are heterozygous in the bulk genomic DNA sample.
Therefore I expect about equal numbers of reference and alternate alleles from the libraries from 1/6th of cells.
Data were 75bp fragment reads from ABI SOLiD5 converted to base space with exact call chemistry. Reads were mapped with BWA and SNP called using the GATK pipeline. Coverage at SNP loci was about 1.6x.
I find I have a 2.08:1 ratio of reference to alternate alleles instead of the 1:1 ratio.
I have checked if the ratio is related to low vcf genotype quality score or coverage and the ratio appears constant irrespective of either factor.
In the bulk DNA sample the ratio of reference to alternate reads at heterozygous positions was 1:1 suggesting only a slight bias in favour of calling reference alleles from reads.
Any suggestions about what might be going on?
Therefore I expect about equal numbers of reference and alternate alleles from the libraries from 1/6th of cells.
Data were 75bp fragment reads from ABI SOLiD5 converted to base space with exact call chemistry. Reads were mapped with BWA and SNP called using the GATK pipeline. Coverage at SNP loci was about 1.6x.
I find I have a 2.08:1 ratio of reference to alternate alleles instead of the 1:1 ratio.
I have checked if the ratio is related to low vcf genotype quality score or coverage and the ratio appears constant irrespective of either factor.
In the bulk DNA sample the ratio of reference to alternate reads at heterozygous positions was 1:1 suggesting only a slight bias in favour of calling reference alleles from reads.
Any suggestions about what might be going on?
