Mascot is the name given to the software developed by Matrix Science for proteomic data analysis. Generally, to identify an unknown protein the target is subjected to protease treatment and the peptide mass fingerprint is recorded by a mass spectrometer (MALDI or ESI/MS). The peptide masses are then used to perform a pattern search in all available databases. Mascot does the job of matching a list of given masses to the MSDB (Mass Spec Database), specifically designed for peptide mass searches.
The Mascot search engine software comes with a complete guide to installation and operation. Our department is running it as a web service enabling the entire campus to benefit, but it can also be operated in the stand-alone mode. For our applications, the software is currently being run on a Pentium IV 3.2 GHz machine with 2 GB of RAM under Microsoft Windows 2000 Pro, however, the software is available for both Windows and Linux platforms. The amount of hard disk space used depends on the databases being maintained. An 80 GB hard disk drive is generally sufficient if the software is being run on an in-house server with only a few databases. The software initially takes some time to install and run, but once loaded, it can be run completely unattended. The system comes with its own web pages that can be accessed over the internet and the database query is submitted through a web browser. The software has been scripted in Perl, provided on the disc for installation.
Three types of queries can be performed with this software: 1) PMF search, 2) Sequence query and 3) the MS/MS ion search (we generally use the PMF and MS/MS ion searches). Options available on the web page for submitting the job include taxonomy (which is useful if one knows the source of the protein), protease used to digest the protein, modifications made, etc. Moreover, mass tolerance is another criterion to increase the stringency in the search. The result of the search appears in another web page which automatically gives a short description of the protein and offers the option of a longer detail analysis of the protein that is matched.
Mascot software comes by default with the MSDB database during software installation. It gives an option for adding databases other than MSDB (e.g. NCBInr, Trembl, Swissprot, etc.) with instructions for uploading in the installation manual. However, to add these databases one will have to download the necessary FASTA files from the FTP servers for the respective databases. In addition, other databases can be compiled, provided that the relevant FASTA files are available. For example, within a specific bacterial species a separate database can be generated if its FASTA file is available. We have entire Mycobacterium tuberculosis and Mycobacterium smegmatis databases in our server.
The software has an option of loading the database in the physical memory (RAM) so that the search can be speeded up. It is therefore advisable to have a high amount of RAM, so that frequently used databases can be locked in memory, improving search speeds. Another benefit with this software is that the scripts for updating the database are already available in the software and have been explained elaborately in the manual. Even without knowledge of Perl (the coding language used for the scripts), these scripts can be modified as desired. Finally, the software keeps a log entry of all searches performed, including the IP addresses of related job submissions. The results of the search are saved and can be obtained later. It also has an option of sending results by e-mail to the users who have submitted the job.
Vikas Jain
Research Scholar
Molecular Biophysics Unit
Indian Institute of Science