Description
PrabHot is a webserver which can effectively identify hot spots on protein-RNA binding interfaces using an ensemble approach. Figure 1 shows the flowchart of PrabHot. A reference dataset of 47 protein-RNA crystal structures is generated from literature curation and Barik et al.'s work (Barik et al., 2016). PrabHot calculates four primary sources of information, namely network, exposure, sequence and structure determinants. The optimal features are selected using the Boruta feature selection algorithm. Then three hot spot classifiers, include SVM (Support Vector Machine), GTB (Gradient Tree Boosting) and ERT (Extremely Randomized Trees), are integrated using a voting approach (EVC). Finally, the performance is evaluated on the benchmark dataset and the independent test dataset.
Current Version
* Current Version Number: 1.2 * Release Date of Current Version: 10/30/2018
Contact Info
E-mail:leideng(at)csu.edu.cn
References [1] Yuliang Pan, Zixiang Wang, Weihua Zhan and Lei Deng*. Computational identification of binding energy hot spots in protein-RNA complexes using an ensemble approach. Bioinformatics, 2017, doi: 10.1093/bioinformatics/btx822
