Proasis2 - Protein Structure Database and Visualisation System

Proasis2 is a protein structure database and visualisation system for drug discovery developed in collaboration with several major pharmaceutical companies.

How Proasis2 Works

• Protein structures important to medicinal chemistry projects are stored in a relational database
• Using a web interface, researchers can search and retrieve structures, and launch a graphics package displaying a structure with the click of a button. A range of viewing modes is available, showing the entire protein or just the active site, highlighting key structural features. Multiple structures can also be superimposed
• Proasis2 provides chemists, crystallographers, and modellers with quick and easy to use tools to explore binding sites and obtain a deeper understanding of the factors determining tight binding. The project based interface gives a distinct 'medicinal chemistry' view of the data
• For crystallographers, the database enables protein structures to be readily stored, annotated, and retrieved. The database administration interface enables database contents to be easily updated
• For modellers, the system removes much of the burden of handling protein structure data and automates many time-consuming tasks

Key Features

• Easy to customise
• Storage of in-house, public domain, X-Ray, NMR and modelled structures in a relational database
• Intranet based
• Focus on ligands in binding sites
• Storage of a wide range of data including crystallographic data, ligand data, sequence data and project specific data
• Multiple binding sites can be defined for any protein chain, and difficult protein-ligand complexes, such as multi-liganded heterodimers, can be fully handled
• For every known structure, multiple pre-defined protein structure viewing modes are available
• Explores non-bonded interactions in a binding site using novel methods
• Overlays multiple structures at the click of a button
• Overlays can be performed based on sequences, specific residues, ligand substructures, and ligand similarity
• In-house projects can be viewed as a hierarchical tree, and multiple hierarchical classification schemes can be defined
• Tools are available for the automatic updating of multiple classification schemes (eg, Kinase and EC Schemes)
• Advanced structure, substructure, text, sequence and recent update searching is available
• Easy to highlight protein sequences showing specific contacts with ligands and other small molecules
• Storage and visualisation of electron density maps
• Symmetry module for building and displaying symmetry molecules, molecular assemblies and crystal packing arrangements
• Extensive database administration facilities
• Automated methods available for batch loading structures into the database
• Explores all monomer units in oligomeric systems
• Manages both liganded and un-liganded structures
• Able to store supplementary data files, reports, presentations, images, etc.
• Comprehensively manages all small molecules associated with a protein structure
• Robust backing-up using xml
• Source code available
• A wide range of browsers and graphics packages (PyMol, Jmol, AstexViewer2, Chime, Rasmol, DSViewer Pro) supported
• Secure
• Competitive entry and maintenance costs
• Public domain structures for almost all known protein-ligand complexes have been characterised and deposited into the system

Proasis2 makes the most valuable and frequently used molecular modelling tasks as routine as searching the internet, for example. It empowers chemists to explore their own ideas and it can free up modellers' time so that they can concentrate on the more sophisticated problems in drug discovery.

Proasis2 Product Brochure (.doc).

For more information, please view slides here from a recent presentation on Proasis2.

The latest update of Proasis2, version 2009, was released Sep. 2009. A 90-day trial version is available upon request.

For research organisations based in Japan, please contact Mr.Syuinichi Ozawa, Infocom Corporation, Bioscience Department, Life Science Division, Mitsui-Kaijou Annex, 3-11 Kanda-Surugadai, Chiyoda-ku, Tokyo, 101-0062, Japan, e-mail: s.ozawa@infocom.co.jp

ViewContacts - Identify Protein-Ligand Interactions

ViewContacts is software for helping chemists, modelers, and crystallographers gain a better understanding of the non-covalent interactions in protein-ligand complexes.

ViewContacts identifies all commonly occurring interactions types, including hydrogen bonding, ionic pairs, and van der Waals contacts, and many, less well understood types, including:

• cation-pi interactions
• hydrogen bonding to pi-systems
• halogen bonding
• orthogonal dipolar alignment
• dipolar antiperiplaner interactions
• pi-stacking
• pi edge-to-face contacts
• hydrogen bonding involving CH groups

An important component of ViewContacts is the handling of solvation. It enables the easy identification of unfavorable interactions due to desolvation, the separation of buried and solvated contacts, the viewing of water-mediated protein-ligand hydrogen bonds, and the ranking of binding site water molecules.

ViewContacts was developed in collaboration with one of the world's leading pharmaceutical research companies, F. Hoffmann La-Roche, Basel, Switzerland, over several years. The software includes the following features:

• Incorporates many recent developments from structural biology, molecular modeling, molecular recognition, and database mining studies
• Built/parameterized using highly curated version of the latest data sets from the PDB
• Carefully handles bound water molecules
• Identifies unfavorable contact pairs
• Identifies potential desolvation penalties
• Creates PyMol scripts for fast visualization
• Easily linked with Proasis2

The application is a stand-alone command line tool that runs under Linux. For more information, please see slides from a recent presentation on ViewContacts

A free evaluation is available upon request.

Spinifex - Chemical Similarity Searching

Spinifex is advanced chemical similarity software that uses a graph-based approach. Spinifex rapidly computes the Maximum Overlapping Set (MOS) of a pair of chemical structures. The MOS is a generalisation of the Maximum Common Subgraph (MCS) and can be thought of as the largest set of substructures that two compounds share. Molecular comparisons based on the MOS are very useful for clustering chemically related compounds [1].

Spinifex has been fine-tuned specifically for handling molecules relevant to medicinal chemistry. The methods used for handling rings, for differentiating atom types, for measuring similarity, and for generating meaningful clusters, have been optimised to meet the needs of HTS data analysis, lead discovery and lead optimisation.

The application is a command line tool that runs under Linux. It can be executed on a single processor computer or in parallel, distributed over multiple computers using PVM. Also available is a Web based tool for HTS Data analysis which automatically runs Spinifex and clustering software from a web page and loads output into a spreadsheet.

In additional to chemical graph matching, Spinifex includes software for: creating molecular alignments, hierarchical clustering, viewing multiple molecules in PyMol and RasMol, and utilities for managing and manipulating sdf files.

For more information, please view slides here from a recent presentation on Spinifex.

The latest update of Spinifex, version 2.9, was released May 2008. A free evaluation version is available upon request.

[1] 'A Robust Clustering Method for Chemical Structures', M. Stahl, H. Mauser, M. Tsui, N.R. Taylor, J. Med. Chem, 2005, 48 (13), 4358-4366.

Please contact us for more information info@desertsci.com