X-SITE v.1.0

Description

X-SITE is an empirically-based method for predicting favourable interaction sites for different atom types at the surface of a protein, such as at a dimer interface, or at a molecular-recognition or binding site. It is particularly useful for docking studies and in protein ligand design.

The empirical data on which the X-SITE predictions are based are the directional atomic contacts observed in high-resolution protein structures (eg different atoms types have different 3D contact preferences relative to each of the 20 amino acid sidechains found in proteins).

The 3D distributions are extracted for different probe atom types around different 3-atom fragments. They can be applied, say, to the sidechains making up the surface of a binding site to predict the favourable regions for different probe atom types within the binding site.

Documentation

Outputs

The X-SITE predictions can be viewed using a number of different molecular graphics packages, or plotted as PostScript or rendered 3D images:-

• AVS - (Advanced Visual Systems Inc., Waltham, MA, USA)
• CCP4 - (Collaborative Computational Project, Number 4, 1994)
• FRODO - (Jones, 1978)
• Insight II - (Molecular Simulations Inc., Burlington, MA, USA)
• O - (Jones et al., 1991)
• PostScript - (Adobe Systems Inc., 1985)
• QUANTA - (Molecular Simulations Inc., Burlington, MA, USA)
• RASMOL - (Sayle & Milner-White, 1995)
• Raster3D - (Bacon & Anderson, 1988; Merritt & Murphy, 1994 )
• SYBYL - (Tripos Associates, Inc., St Louis, MO, USA)

Availability

Source code, documentation and related data file can be picked up from the directory:-

pub/xsite

The source code is concatenated into a single, encrypted tar file called xsite.tar.Z.enc.

Users must sign a Confidentiality Agreement, supplied with the source code, and return it to the author to get the decryption key. The data files and documentation are not encrypted.

Further details, including up-to-date release details, can be found in the README file.