[EMBOSS] pepstats: (im)probability of expression in inclusion bodies?

[Fernan Aguero]

Hi,

I was just looking at the output of pepstats for the
probability of expression in inclusion bodies in E.  coli,
and noticed that in some cases the values reported are not
for the probability but for the 'Improbability' of
expression.

Why is that? I have just run one small genome (805 genes)
through pepstats and I've found 625 cases of 'Improbability'
vs 180 for 'Probability'.

Suspiciously, in both cases the values ranged from ~ 0.49 to
~ 0.97.  Am I safe to assume that pepstats is reporting low
probability values of expression in inclusion bodies as high
Improbability values instead? (i.e. 1 - probabiliy )? 

If this is the case, why? Why not standardize the output for
all cases, reporting just the probability values?

This is on EMBOSS-5.0.0

BTW, the fine manual does not say anything about this
behaviour :)

Fernan