adding documentation that let the user know what indexes each component of the PDB files need to be.

[tsantaloci]

For example, the res_num needs too be between spaces resnum = int( line[22:26] ) (A residue number cant be greater than 4 digits with this notation which may be a problem for proteins with larger than 100,000 residues) and chain needs too be chain = line[21].

[natedey]

The largest known protein only has 35000 residues 🤣
But we were having problems with element names w two characters coming from AMBER MD and having the second letter chopped off by RINRUS (changing FE to F which caused a lot of chaos in the cP450 project)

In general, PDB formatting is very rigid. We would expect the user to carefully adhere to the formatting. We could refer them to format validation tools, or we could incorporate some kind of sanity check for the most important things that could break RINRUS?

format validation tools:
http://www.csb.yale.edu/userguides/datamanip/pdb_validation/pdbval_descrip.html
https://validate-rcsb-1.wwpdb.org/

[natedey]

Ahh yikes, this is the same situation as in #38. I was silly to not realize that res ids can easily exceed 9999 if you have a big box of water around your protein from an MD simulation. Bachega has written a script to fix the problem that occurs when saving PDBs in VMD.

We definitely need to warn the users of common cases where a PDB format can go out of bounds w RINRUS.

[natedey]

@tsuhagia and @ferbachega each have made utilities to handle some problem cases.
TJ has a script to unpack crystal space group symmetry from older PDB files
TJ and Bachega both have scripts to handle VMD and AMBER PDB output from MD where there are over 10000 fragments due to protein solvation. We may need to create a "utilities" directory. Then we need to decide whose > 10000 fragment script to package with RINRUS.

[natedey]

I'm not sure I ever realized that there are specific justification guidelines for PDB character fields too!

[natedey]

@tsuhagia and @ferbachega should send me the scripts they made to fix > 10000 fragment issues and these should be included with RINRUS. We should begin a list of common pdb problems and edge cases.