Szymoszek, Andrzej;Zacharias, Martin - Structure prediction of protein segments with low target/template sequence identity based on a reduced protein model

ECCB 2002 Poster sorted by: Author | Number

Next | Previous poster (in order of the view you have selected)

Title: Structure prediction of protein segments with low target/template sequence identity based on a reduced protein model	P160
Szymoszek, Andrzej; Zacharias, Martin andrzej@imb-jena.de Jena Centre for Bioinformatics, Institut fuer Molekulare Biotechnologie, Beutenbergstr. 11, D-07743 Jena

Modelling of the three-dimensional structure of a protein based on sequence similarity to a protein with known structure, is limited by a degree of sequence identity. Frequently, the quality of the target- template sequence alignment is non-uniform along the sequence: parts of the target sequence differ strongly from the template, whereas modelling of other parts based on the template structure is possible with a high confidence. In many cases, the former regions are larger than small loop regions, that can be modelled with help of loop structure data bases. In order to rapidly generate and evaluate a large number of possible conformations for a protein segment, we have designed a reduced model with two pseudo atoms per amino acid. The model allows a very rapid generation of protein segment conformations compatible with the boundaries given by the part of the protein, that can be accurately modelled based on the template structure. The structure generation is achieved by a combination of known structural motifs, and secondary structure elements, providing a variety of putative protein segment conformations. The interaction between amino acids is described by a soft Lennard-Jones type energy function, parameterised according to the concept of inter-residue contact energies, extracted from folded protein structures [1]. The aim of the approach is to eliminate a large number of putative protein segment conformations at the level of the reduced protein model, prior to a further evaluation at the atomic resolution. The method has been applied to a number of test systems with known structures. Furthermore, its application to the short chain dehygrogenase reductase (SDR) family of enzymes, important for steroid metabolism, will be presented. The target 3D structures in this case are unknown, and the sequence similarity to known templates is very limited.

[1] Miyazawa, S., Jernigan, R.L., Self- Consistent Estimation of Inter-Residue Protein Contact Energies Based on an Equilibrium Mixture Approximation of Resides, PROTEINS: Structure, Function and Genetics 34:49-68 (1999).