FragHMMent: A computational tool for predicting residue-residue contacts

FragHMMent is a computational tool for predicting residue-residue contacts from amino acid sequence using multi-data hidden Markov models trained on local neighborhoods (local descriptors of protein structure). It is written in Java. The program takes as an input one or more sequences in fasta-format and outputs results in the CASP RR format.

Download the program and follow the instructions in the readme file.

Server implementation of the program

P. Björkholm, P. Daniluk, A. Kryshtafovych, K. Fidelis, R. Andersson and T. R. Hvidsten. Using multi-data hidden Markov models trained on local neighborhoods of protein structure to predict residue-residue contacts. Bioinformatics 25: 1264-1270, 2009.

Questions: pbh@sbc.su.se

For more information about local descriptors, see:
T. R. Hvidsten, A. Kryshtafovych and K. Fidelis. Local descriptors of protein structure: A systematic analysis of the sequence-structure relationship in proteins using short- and long-range interactions, Proteins: Structure, Function, and Bioinformatics 75 (4): 870-884, 2009.

More papers on local descriptors here.

Mirrors