Search

Identification of B-cell epitopes (BCEs) plays an essential role in the development of peptide vaccines and immuno-diagnostic reagents, as well as antibody design and production. In this work, we generated a large benchmark dataset comprising 124,879 experimentally supported linear epitope-containing regions in 3567 protein clusters from over 1.3 million B cell assays. Analysis of this curated dataset showed large pathogen diversity covering 176 different families. The accuracy in linear BCE prediction was found to strongly vary with different features, while all sequence-derived and structural features were informative. To search more efficient and interpretive feature representations, a ten-layer deep learning framework for linear BCE prediction, namely NetBCE, was developed. NetBCE achieved high accuracy and robust performance with the average area under the curve (AUC) value of 0.8455 in five-fold cross-validation through automatically learning the informative classification features. NetBCE substantially outperformed the conventional machine learning algorithms and other tools, with more than 22.06% improvement of AUC value compared to other tools using an independent dataset. Through investigating the output of important network modules in NetBCE, epitopes and non-epitopes tended to be presented in distinct regions with efficient feature representation along the network layer hierarchy. The NetBCE is freely available at https://github.com/bsml320/NetBCE.

As an important protein acylation modification, lysine succinylation (Ksucc) is involved in diverse biological processes, and participates in human tumorigenesis. Here, we collected 26,243 non-redundant known Ksucc sites from 13 species as the benchmark data set, combined 10 types of informative features, and implemented a hybrid-learning architecture by integrating deep-learning and conventional machine-learning algorithms into a single framework. We constructed a new tool named HybridSucc, which achieved area under curve (AUC) values of 0.885 and 0.952 for general and human-specific prediction of Ksucc sites, respectively. In comparison, the accuracy of HybridSucc was 17.84%–50.62% better than that of other existing tools. Using HybridSucc, we conducted a proteome-wide prediction and prioritized 370 cancer mutations that change Ksucc states of 218 important proteins, including PKM2, SHMT2, and IDH2. We not only developed a high-profile tool for predicting Ksucc sites, but also generated useful candidates for further experimental consideration. The online service of HybridSucc can be freely accessed for academic research at http://hybridsucc.biocuckoo.org/.

In eukaryotes, protein phosphorylation is specifically catalyzed by numerous protein kinases (PKs), faithfully orchestrates various biological processes, and reversibly determines cellular dynamics and plasticity. Here we report an updated algorithm of Group-based Prediction System (GPS) 5.0 to improve the performance for predicting kinase-specific phosphorylation sites (p-sites). Two novel methods, position weight determination (PWD) and scoring matrix optimization (SMO), were developed. Compared with other existing tools, GPS 5.0 exhibits a highly competitive accuracy. Besides serine/threonine or tyrosine kinases, GPS 5.0 also supports the prediction of dual-specificity kinase-specific p-sites. In the classical module of GPS 5.0, 617 individual predictors were constructed for predicting p-sites of 479 human PKs. To extend the application of GPS 5.0, a species-specific module was implemented to predict kinase-specific p-sites for 44,795 PKs in 161 eukaryotes. The online service and local packages of GPS 5.0 are freely available for academic research at http://gps.biocuckoo.cn.

Various posttranslational modifications (PTMs) participate in nearly all aspects of biological processes by regulating protein functions, and aberrant states of PTMs are frequently implicated in human diseases. Therefore, an integral resource of PTM–disease associations (PDAs) would be a great help for both academic research and clinical use. In this work, we reported PTMD, a well-curated database containing PTMs that are associated with human diseases. We manually collected 1950 known PDAs in 749 proteins for 23 types of PTMs and 275 types of diseases from the literature. Database analyses show that phosphorylation has the largest number of disease associations, whereas neurologic diseases have the largest number of PTM associations. We classified all known PDAs into six classes according to the PTM status in diseases and demonstrated that the upregulation and presence of PTM events account for a predominant proportion of disease-associated PTM events. By reconstructing a disease–gene network, we observed that breast cancers have the largest number of associated PTMs and AKT1 has the largest number of PTMs connected to diseases. Finally, the PTMD database was developed with detailed annotations and can be a useful resource for further analyzing the relations between PTMs and human diseases. PTMD is freely accessible at http://ptmd.biocuckoo.org. Keywords: Posttranslational modification, Phosphorylation, PTM–disease association, Disease–gene network, AKT1