Predicting
disulfide connectivity from protein sequence
Advanced
Computational Modelling Centre,
The University of Queensland
If you find this service useful, please cite:
Jiangning Song, Zheng Yuan, Hao Tan, Thomas Huber and Kevin Burrage
Predicting disulfide connectivity from protein sequence using multiple
sequence feature vectors and secondary structure
Bioinformatics. 2007, 23(23):3147-3154.
The
input to the service includes three fields.
1.
Email address:
Input the email address to which the result should be sent. Please type carefully and check the email address you entered, since results will not reach you if you mis-typed your address.
2.
The SVR models of choice:
We used two datasets to build the SVR models: SP39
dataset and SP39 template
dataset. Please note that using different SVR models can lead to
different prediction results of disulfide connectivity patterns. These
two SVR models are provided for readers' reference and cross-validation.
The sequence encoding scheme utilized here is based on PSI-BLAST profiles
in the form of PSSMs, predicted secondary structure information by PSIPRED,
twenty amino acid composition, the logarithmic DOC value, the cysteine
ordering, the normalized protein molecular weight and the normalized
sequence length.
3.
Protein sequence:
Simply input the amino acid sequence of the query protein in one-letter
format like the sample sequence given below. Protein sequences with more than 12 cysteine residues will not be processed!
Email address:
Please select
the following SVR models to predict disulfide connectivity pattern:
SVR-SP39 (SVR model built on SP39 dataset)
SVR-SP39-template (SVR model built on SP39-template dataset)
|