Transcriptional strand bias on SNV from exome data

Question

[Chirag Nepal]

Hi there,

I want to find SNV that show bias in transcribed/untranscribed strand

Here is one SNV from varscan, with tumor/normal read counts, with header and data

chrom   position    ref var normal_reads1   normal_reads2   normal_var_freq normal_gt   tumor_reads1    tumor_reads2    tumor_var_freq  tumor_gt    somatic_status  variant_p_value somatic_p_value tumor_reads1_plus   tumor_reads1_minus  tumor_reads2_plus   tumor_reads2_minus  normal_reads1_plus  normal_reads1_minus normal_reads2_plus  normal_reads2_minus

chr11   1092459 C   A   300 30  9.09%   C   245 71  22.47%  M   Somatic 1.0 1.9447764810399614E-6   119 126 51  20  156 144 21  9

I calculated ratio of plus_strand/minus_strand for tumor/normal variant

cat filename | awk 'BEGIN {OFS="\t"} { print $0, $16/$17, $18/$19, $20/$21, $22/$23 }'


chr11   1092459 C   A   300 30  9.09%   C   245 71  22.47%  M   Somatic 1.0 1.9447764810399614E-6   119 126 51  20  156 144 21  9   0.944444    2.55    1.08333 2.33333

Tumor_ref_allele is 0.94 between plus/minus_Strand

Tumor_var_allele is 2.55 between plus/minus_Strand [indicate preference of mutation in -ve strand]

Normal_ref_allele is 1.08 between plus/minus_Strand

Normal_var_allele is 2.33 between plus/minus_Strand [indicate preference of mutation in -ve strand]

I looked at the position, chr11-1092459 is MUC2 gene on plus strand. Is it correct to say that variant_allele is biased in non-transcribed strand ?

Any suggestions are welcome !

Thanks !

[Brian Bushnell]

It doesn't seem like this analysis would be productive in unstranded shotgun DNA data generated from double-stranded DNA (particularly if it is amplified). Can you describe your library preparation in more detail?