fcyu / pipi Goto Github PK

I have only one question: why such an excellent paper was not published in Nature?

Proteins sequences present in Fasta files may include -not too rarely- undetermined amino-acids (often marked as * ).
It seems PIPI does not behave well when encountering a symbol in a protein sequence that is not described in its own internal amino-acid list, and crashes as reported below.

It would probably better if the software could just at least remove this specific protein from the database, warns the user, and pursue the rest of the processing.

It seems a Post-Translational-Modification file is needed to run PIPI.

This file is however missing in PIPI-1.2.8 release -leading to PIPI crash-, and no mention is made of it or any instruction given in the Readme to build your own.

The file can however be found in GitHub sources.

The parameter file provided in the 1.2.8 release of PIPI is missing the following lines that are present in the source code version of the same file:

min_potential_charge = 2
max_potential_charge = 4

Here is the detailed error obtained running the software:

Searching the source code reveals the error being due to missing parameters; adding them worked.
Not sure if the software is intended to work with default values if not provided, but in the current case it just throws back an exception related to the unfound values.