PAT: a protein analysis toolkit for integrated biocomputing on the web

Abstract

PAT, for Protein Analysis Toolkit, is an integrated biocomputing server. The main goal of its design was to facilitate the combination of different processing tools for complex protein analyses and to simplify the automation of repetitive tasks. The PAT server provides a standardized web interface to a wide range of protein analysis tools. It is designed as a streamlined analysis environment that implements many features which strongly simplify studies dealing with protein sequences and structures and improve productivity. PAT is able to read and write data in many bioinformatics formats and to create any desired pipeline by seamlessly sending the output of a tool to the input of another tool. PAT can retrieve protein entries from identifier-based queries by using pre-computed database indexes. Users can easily formulate complex queries combining different analysis tools with few mouse clicks, or via a dedicated macro language, and a web session manager provides direct access to any temporary file generated during the user session. PAT is freely accessible on the Internet at http://pat.cbs.cnrs.fr.

Figure 1

PAT interface to CLUSTALW. This example highlights several PAT features: The ‘Input’ area shows how several input data formats can be mixed together in a single query and automatically recognized and reformatted by PAT for processing. Several successive redirections of the CLUSTALW output can be seen, first to PREDATOR and DSC, then to CONSENSUS, and finally to the COLOR HTML display shown in Figure 3.

Figure 2

Overall processing flowchart corresponding to one web query. The main web forms in the ‘Tool’ page are shown with bold letters. The redirections displayed on a gray background are optional and could be repeated using different tools if the user wishes to create a longer pipeline.

Figure 3

Final colored HTML output of the pipeline shown in Figure 1. Protein sequences, secondary structure prediction and consensus are displayed using the CLUSTALW alignment and specific color schemes.

Figure 4

A session example corresponding to the execution of a simple macro. The data input consists of two protein structures (2eti and 4cpaI) and the macro requires successive execution of CE (structural alignment), ProFit (structure superimposition) and Jmol (display of superimposed structures). Clicking the O1, O2 or O3 links in the output table display results of each elementary step in separate windows, respectively (blue arrows).