disulfinder(1) cysteines disulfide bonding state and connectivity predictor


disulfinder [OPTIONS]


'disulfinder' is for predicting the disulfide bonding state of cysteines and their disulfide connectivity starting from sequence alone. Disulfide bridges play a major role in the stabilization of the folding process for several proteins. Prediction of disulfide bridges from sequence alone is therefore useful for the study of structural and functional properties of specific proteins. In addition, knowledge about the disulfide bonding state of cysteines may help the experimental structure determination process and may be useful in other genomic annotation tasks. 'disulfinder' predicts disulfide patterns in two computational stages: (1) the disulfide bonding state of each cysteine is predicted by a BRNN-SVM binary classifier; (2) cysteines that are known to participate in the formation of bridges are paired by a Recursive Neural Network to obtain a connectivity pattern.


A. Ceroni, A. Passerini, A. Vullo and P. Frasconi. DISULFIND: a Disulfide Bonding State and Cysteine Connectivity Prediction Server, Nucleic Acids Research, 34(Web Server issue):W177-W181, 2006.

For the disulphide connectivity predictor see:

A. Vullo and P. Frasconi. Disulfide Connectivity Prediction using Recursive Neural Networks and Evolutionary Information, Bioinformatics, 20, 653-659, 2004.

For the cystein bonding state predictor see:

P. Frasconi, A. Passerini, and A. Vullo. A Two-Stage SVM Architecture for Predicting the Disulfide Bonding State of Cysteines, Proc. IEEE Workshop on Neural Networks for Signal Processing, pp.25-34, 2002.
A.Ceroni, P.Frasconi, A.Passerini and A.Vullo. Predicting the Disulfide Bonding State of Cysteines with Combinations of Kernel Machines, Journal of VLSI Signal Processing, 35, 287-295, 2003.


-a, --alternatives=NUMBER
alternative connectivity patterns (default=3)
-o, --output=DIR
output dir where predictions will be saved (default=$PWD)
-p, --psi2=FILE|DIR
input in psi2 format (PSI-BLAST Matrix in ASCII), either a single file or a directory(?). Generate this with "blastpgp -j <N> -Q FILE" where N >= 2.
-r, --rootdir=DIR
work directory (default=~/disulfinder)
-k, --pkgdatadir=DIR
package data directory containing Models (default=/usr/share/disulfinder)
-F, --format={html|ascii}
output format type (default=ascii)
-d --blastdb=DIR
blastpgp -d option (default=/data/sp+trembl)
-c, --cleanpred
cleanup intermediate prediction files (default=false)
-P, --usepssm
use pssm instead of counts for profiles (default=false)
-C, --knownbondingstate
assume bonding state is known (one file for each chain in directory <rootdir>/Predictions/Bondstate/Viterbi) (default=false)
-v, --version
disulfinder version
-?, --help
help screen


"disulfinder -a 1 -p /usr/share/doc/disulfinder/examples/res_id_41483.blastPsiMatTmb -o ./disulfinder_results_dir"


default package data directory
default work directory


The work directory is not cleaned up automatically.


Ceroni A, Passerini A, Vullo A, Frasconi P.

Packaging by Laszlo Kajan <[email protected]>