pKa Database Documentation

Implementation Notes

• • Database server runs as an index server of computing server.

  • A special file in working directory instructs database server to compile the calculation to the database. This file contains submitter, publish flag (to publish, private, or delete), Method flag (MCCE, other, experimental) etc.

  • A script is provided to compile/delete a calculation to database on demand.

  • Database re-scan computing server daily.

  • Pseudo working directory for experimental values.

  • Search and refine method to narrow down search results.

  • MCCE calculation is presented by MFE analysis interface.

How to use Database

How does it work?

Database server and computing server connection

Database file

Submit script

Automatic scan

Pseudo working directories for pKas obtained from other methods

How to prepare a submit file and convert a working directory into database?

Prerequisites:

1. Energy opp files are in energies directory. Some versions of MCCE have this in a compressed file.

2. The working directory name is in pdb file name, and it may have extension and other parts after "." For example these are recognizable working directory names:

3. 1a2p, 1a2p.a, 1a2p.old

Compile:

Run dbprep.py 

This will convert all energy terms in to a human readable file with key value pairs for further process into mysql file. It also calculates mean field energy terms to approximate the composition of pKa values of each residues. Protein information is obtained from PDB website RESTful service.

Edit pka.sub if necessary

Run pka2sql.py

This converts the file pka.sub into pka.sql. This file is in sql format that contains the same information as pka.sub.

Load to mysql:

Copy pka.sql to database server, run "mysql -u mcce -p -Dpkadatabase < pka.sql"