| |
CEMC - Multiple Protein Structure Alignment Server
FAQs on CE-MC Webserver
1. What's a JOB ID?
* A JOB ID is an unique identifier created for each job request on the CE-MC server. All input and output data for each requested job are written to a sub-directory with its name same as the JOB ID. Alignment results are stored on the CE-MC server for 72 hourse and users could retrieve their results using JOB ID.
2. How does CE-MC server generates multiple alignments ?
* CE-MC program runs based on two independent algorithms, Combinatorial Extension (CE) and Monte Carlo (MC) optimization methods. CE-MC program performs the following steps in the given order.
1. Parses structure data files and stores component data as POM database objects
2. Generates pair-wise alignments between chains in all-to-all combination using CE program
3. MC program generates a seed alignment from CE pair-wise data and optimizes the multiple alignments using MC optimization method
4. CE-MC output (text format) is reformatted to JOY/html format using JOY program
3. Why can't I get my alignments right away? Why there are jobs ahead of my job?
* CE-MC alignments usually time consuming. In order to keep the server up and running for everyone, we adopted a scheduling system to process jobs in the order they were requested. Results are sent to the email address provided upon completion of the job.
4. How long does it take for processing of my job request ?
* Assuming that there are no other jobs in queue ahead of your job request, the net processing time for running a job request depends primarily on the number of protein chains being aligned, and also the length of protein chains and the structural diversity of the family being aligned. The computational complexity of CE-MC algorithm is 'quadratic'. Approximate time estimations depending on the number of chains are: 4 chains, 3 minutes; 10 chains, 10 minutes; 20 chains, 25 minutes etc.
5. What is a Master structure ?
* Master structure is used as a reference, while building the seed alignment before Monte Carlo optimization. Master structure is chosen as the one containing the higherst average Z-score calculated from the Z-score values of all pair-wise alignments for each chain. Flanking end fragments in the slave structures that are not in alignment with the master structure, are truncated and not represented in the multiple alignment.
6. How can I get this program running on my local machine?
* If you plan to use this method extensively, we encourage you to download it and install on your local machine.
|
|
