V-phaser Strand bias p-value >1

[Konstantin Chukreev]

Hello!

I ran V-phaser for my .bam file and got very strange results (here's some of the lines from .fdr.var.txt file):

# Ref_Pos       Var     Cons    Strd_bias_pval  Type    Var_perc        SNP_or_LP_Profile

# ------------------------------------------------------------

648505  T       C       1.351   snp     1.942   C:55:46 T:1:1

648507  G       C       1.351   snp     2.02    C:55:42 G:1:1

655550  T       C       1.209   snp     1.626   C:66:55 T:1:1

655598  T       C       1.209   snp     1.639   C:65:55 T:1:1

1239960 G       T       1.274   snp     2.174   G:1:1   T:48:42

As far as I get the number going right after the letters is strand bias p-value. But somehow this value is more than 1. I don't have any ideas how could it happens. And also there are some strange p-values connected with quite good variants:

590988  C       T       0.5827  snp     34.78   C:24:16 T:34:41

591025  A       C       0.5563  snp     45.52   A:32:29 C:31:42

Can you tell me, is it a bug or I just misunderstand meaning of the numbers?

Sincerely,

Konstantin.