Does anyone know of a command-line (non-GUI) tool like SCWRL that will optimize the side-chains of a protein structure, but allow you to somehow specify that you want not the most optimal rotamer, but rather the 2nd or 3rd or 4th most optimal rotamer? My system is a static system, but I need to look at the effects of changing a rotamer in several hundred proteins.
I know this can be done in Chimera, it will give you a list of rotamers and their frequency, but I need a command-line tool since it will be part of a larger script that will be used on 100s of files (so I can't just do it manually).
What do you mean by "rotamer" here from the perspective of protein structures ? Is it same as per the definition here http://en.wikipedia.org/wiki/Conformational_isomerism ? If your answer is YES, my answer is valid. You can generate conformers using CRANKITE.
Yes, I am using the same definition of "rotamer," but no, your answer is not valid because I used the word "optimal" several times and cited SCWRL and Chimera to make it clear that I am not looking for random sampling (CRANKITE), I am looking for "optimal"/frequent rotamers (SCWRL/Rotamer libraries).
Most probably it can be done by EGAD a protein design tool, that runs in unix/linux from command line.
I found it a bit difficult to learn and use for protein designing purpose, and finally abandoned it in favor of some other utility at it keeps on crashing for a particular type of protein complex. I hope a simple functionality like single amino acid rotamer optimization would run flawlessly. Have a quick glance at the manual HERE
Since I asked this question I found a useful tool on the PyMOL wiki called Rotamer Toggle. It uses the Dunbrack rotamer library. It may not be the most user-friendly, but I think it's a good programmatic way to handle this problem.
The Metropolis polypeptide backbone conformation sampler. A very short simulation can be run by using a command like
./peptide -r 1000x25 -t 201 d1ctf__.pdb
This will read the initial conformation from file d1ctf__.pdb, perform 25,000 Metropolis moves, while recording a contact map and a snapshot conformation every 1000 steps. The output will have 25 contact maps and 25 snapshots. To get help with more options, use "./peptide -h".
These 25 snapshots will contain different rotamers / conformers you are looking for.
I read the Crankite SCFBM page (1st link you gave) and also looked through the readme for Crankite, and I do not see how I could use it to change just one particular side-chain rotamer. The documentation never even mentions the words "rotamer." Let me know if I'm missing something. Maybe you know of an undocumented feature.
I didn't noticed that the user not used the term 'rotamer'. The paper is about a library for conformational sampling, which is AFAIK same as the rotamers. From the perspective of protein structures rotamers = conformations. CRANKITE is a library to generate conformers. Use the option I have added to the answer not just "2nd or 3rd or 4th most optimal rotamer" you can generate 25 conformers.
Thank you for trying to help, Khader, but I only want to change a side-chain, and only to the next most optimal/frequent/common rotamer. The CRANKITE tool looks good for conformational sampling, but I am not looking for sampling.
To clarify the difference: the command you provided will generate 25 protein conformations, but 1) they will not be ordered according to side-chain frequency and 2) the whole structure and backbone will be different. Technically, I could just run a very long CRANKITE simulation, but then I would have to cluster and sort the snapshots to get the optimal side-chains.
I was in a impression that conformation sampling could be a way to find next optimatal rotamer. I agree with your points if you are specifically looking for specific side-chain rotamers, CRANKITE may not be an ideal option. I am not deleting the answer, hope the discussion will be useful for future reference.
What do you mean by "rotamer" here from the perspective of protein structures ? Is it same as per the definition here http://en.wikipedia.org/wiki/Conformational_isomerism ? If your answer is YES, my answer is valid. You can generate conformers using CRANKITE.
Yes, I am using the same definition of "rotamer," but no, your answer is not valid because I used the word "optimal" several times and cited SCWRL and Chimera to make it clear that I am not looking for random sampling (CRANKITE), I am looking for "optimal"/frequent rotamers (SCWRL/Rotamer libraries).
Please read this section of the page you cited: http://en.wikipedia.org/wiki/Conformational_isomerism#Protein_rotamer_libraries