Autmated Peptide Screening Wiki
Screening peptides for MHC and/or MHC-T cell receptor affinity
Status: Alpha
Brought to you by:
stephanwatkins
Welcome to your wiki!
This is the default page, edit it as you see fit. To add a new page simply reference it within brackets, e.g.: [SamplePage].
The wiki uses Markdown syntax.
Modified version of AutoGrow, autogrow.ucsd.edu/ , with a mutation script added in the class files, and a global change of the defined mutation point from Hydrogen atoms to Calphas, to allow placement of Amino Acids. Deposited are 3 versions, used for my PhD Thesis. These allow for mutation of short 2-15 amino acid peptides, and associated screening of these using AutoDock or AutoDock Vina, allowing for 8 CPU systems to screen 10K peptides a day or 1-2K with flexible residues added in. A detailed method for determining the peptide ligand for any sequenced and modeled T cell receptor will be published and link added at some point. These scripts are written quickly in a couple weeks to meet a task and could be re-written more eloquently. I may or may not work on this project again, the goal would be to introduce amino acids into any protein, written in openCL to allow automated design on GPU's.
Limitations-The standard PDB codes used had to be slightly modified (see fragment library for PDB definitions) which can be parsed on output with pymol to re-write in standard PDB, proline will not insert into angles, random bugs. Peptides should be linear for best results, also changing in the automated scripts for building .pdbqt files to use some/partial flexability helps in Docking peptides.
For bugs, each one has a problem, ideal would be an insert then energy minimization. Bugs range from bad placement of single hydrogen atoms, or reversed coordinates - to + in one xyz point), and prolines being inserted backwards, or one single amino acid not finding a conformer. Still, using one or all of them will generate large libraries of peptides in minutes, and toying you can just screen them as needed, either by stopping and then re-starting the base Autogrow, or submission using a bash shell to autodock or autodock vina.
Included, I appolgise, are random tidbits of things done. All files should be the same name as those in the origional autogrow. In addition there are examples of output generated from some runs under generation_xxx named directories. The standard proceedure for setting up the files is to copy all base java files to a stand alone dir , compile them and then follow the setup for autogrow in the autogrow web page. Good luck.