funDNApred - Fuzzy Cognitive Map approach to DNA residues prediction

funDNApred webserver

The funDNApred webserver predicts residue-level propensities for DNA binding for the input protein sequence. Its predictive model is based Fuzzy Cognitive Map that relies on the information concerning evolutionary conservation (generated with HHblits), putative relative solvent accessibility (generated with ASAquick) and propensity of amino acid types for the DNA binding.

Please follow the three steps below to make predictions:

1. Copy and paste protein sequence(s) into text area

The funDNApred webserver accepts up to 10 FASTA formatted protein sequences with minimum length of 21 residues. Please enter protein sequence(s) in the following text field.

2. Provide your email address (optional)

Please enter your email address in the following text area. A link to results of assessment will be sent to your email address once they are ready. The results will be also available in the browser window.

3. Predict

Click button to launch prediction.

Materials

README file - descriptions of file format for the datasets above
Training dataset - the selected residues, native DNA-binding annotations, RAA values, RSA values, and ECO values for the training dataset
Test dataset - the sequences, native DNA-binding annotations, RAA values, RSA values, and ECO values for the test dataset

References

Upon the usage the users are requested to use the following citation:

Abdollah Amirkhani, Mojtaba Kolahdoozi, Chen Wang, and Lukasz Kurgan. Prediction of DNA-binding residues in local segments of protein sequences with Fuzzy Cognitive Maps. Submitted.

Acknowledgments

We acknowledge with thanks the following software used as a part of this server:

HHBlits - Homology detection by iterative HMM-HMM comparison
ASAquick - A quick neural network-based ASA predictor