ATP-binding sites are valuable drug targets for antibacterial and anti-cancer chemotherapy. Hence, accurately localizing the protein-ATP binding sites is of significant importance for both protein function annotation and drug discovery.
Identifies adenosine triphosphate (ATP)-binding residues from protein sequences thanks to a support vector machine classifier. ATPsite serves for the sequence-based prediction and aims to improve over the predictive quality of the annotation of binging residues, (including sequence alignment and conservation scoring). It can utilize a large set of input features including sequence and predicted structural descriptors.
Predicts adenosine-5’-triphosphate (ATP) specific binding sites. ATPbind employs a combination of several support vector machines (SVMs) based on the random under-sampling technique. It integrates the outputs of two template-based predictors (S-SITE and TM-SITE) with three sequence-based elements: the position specific scoring matrix, the predicted secondary structure, and the predicted solvent accessibility.
Predicts protein- Adenosine-5’ -triphosphate (ATP) binding site. TargetATPsite can report binding pockets from the predicted binding residues with a spatial clustering process. It uses an image sparse representation technique to encode residue evolution information and support vector machines (SVM) to deal with imbalance phenomenon between the positive and negative training samples.