PAPA scores

[sghignone]

hello!
let's say I have a proteome (e.g. fungal, from swissprot), and I scan all entries first with PAPA and the with PLAAC.
By logic, I should expect same PAPA scores in both PAPA and PLAAC output, or I am wrong?
or how should I tweak scripts options to align results?
Thanks for the explanations
SG

[oliverking]

From what I recall the re-implementation of the PAPA algorithm in PLAAC was originally done based on the description of the basic algorithm from Toombs et al. 2010, before the PAPA code and website from Toombs et al. 2012 were released. It is possible that the implementations differ in some details, and if they do then I would regard the versions from Toombs el al. (the Ross group) as the definitive versions of PAPA. Both the Python implementation from the Ross group and the Java implementation in PLAAC have options to dampen the effect of nearby prolines, so that would be one place to check for discrepancies (both to check that the same option is selected, and check on how exactly the dampening is implemented). The command-line version of PLAAC also has options for whether to plot "singly-smoothed" or "doubly-smoothed" tracks for PAPA scores, so that would be another place to check for discrepancies.