STS-NLSP: A Network-Based Label Space Partition Method for Predicting the Specificity of Membrane Transporter Substrates Using a Hybrid Feature of Structural and Semantic Similarity

Abstract

Membrane transport proteins play crucial roles in the pharmacokinetics of substrate drugs, the drug resistance in cancer and are vital to the process of drug discovery, development and anti-cancer therapeutics. However, experimental methods to profile a substrate drug against a panel of transporters to determine its specificity are labor intensive and time consuming. In this article, we aim to develop an in silico multi-label classification approach to predict whether a substrate can specifically recognize one of the 13 categories of drug transporters ranging from ATP-binding cassette to solute carrier families using both structural fingerprints and chemical ontologies information of substrates. The data-driven network-based label space partition (NLSP) method was utilized to construct the model based on a hybrid of similarity-based feature by the integration of 2D fingerprint and semantic similarity. This method builds predictors for each label cluster (possibly intersecting) detected by community detection algorithms and takes union of label sets for a compound as final prediction. NLSP lies into the ensembles of multi-label classifier category in multi-label learning field. We utilized Cramér's V statistics to quantify the label correlations and depicted them via a heatmap. The jackknife tests and iterative stratification based cross-validation method were adopted on a benchmark dataset to evaluate the prediction performance of the proposed models both in multi-label and label-wise manner. Compared with other powerful multi-label methods, ML-kNN, MTSVM, and RAkELd, our multi-label classification model of NLPS-RF (random forest-based NLSP) has proven to be a feasible and effective model, and performed satisfactorily in the predictive task of transporter-substrate specificity. The idea behind NLSP method is intriguing and the power of NLSP remains to be explored for the multi-label learning problems in bioinformatics. The benchmark dataset, intermediate results and python code which can fully reproduce our experiments and results are available at https://github.com/dqwei-lab/STS.

T4SE-XGB: Interpretable Sequence-Based Prediction of Type IV Secreted Effectors Using eXtreme Gradient Boosting Algorithm

“…The combination of different features could depict protein sequences in a more comprehensive manner (Wang et al, 2019b ). As illustrated in Table 1 , using the combined features yield the ACC of 93.95% and the MCC of 0.8346, which are both higher than other PSSM-based features.…”

“…This observation indicates that the features based on PSSM have better performance in the prediction of T4SE when compared with other types of features. The combination of different features could depict protein sequences in a more comprehensive manner (Wang et al, 2019b). As illustrated in Table 1, using the combined features yield the ACC of 93.95% and the MCC of 0.8346, which are both higher than other PSSM-based features.…”

RF-PseU: A Random Forest Predictor for RNA Pseudouridine Sites

“…For the ROC curve, 1-specificity was plotted on the horizontal axis, and sensitivity on the vertical axis. LOO, K-Fold cross-validation, and independent testing are the most widely used methods for predictor evaluation (Mrozek et al, 2015;Cao and Cheng, 2016;Chen et al, 2017Chen et al, , 2018aChen et al, , 2019bPan et al, 2017;He et al, 2018He et al, , 2019Jiang et al, 2018;Xiong et al, 2018;Yu et al, 2018;Zhang et al, 2018;Ding et al, 2019;Feng et al, 2019;Kong and Zhang, 2019;Li and Liu, 2019;Lv et al, 2019a;Manavalan et al, 2019;Shan et al, 2019;Wang et al, 2019a;Wei et al, 2019a,b;Xu et al, 2019;Yu and Dai, 2019). That is the general machine learning evaluation methods (training, validation and testing) are used for optimized model evaluation.…”

Hybrid Semantic Recommender System for Chemical Compounds in Large-Scale Datasets

“…We considered the unrated items as negative ratings, i.e., not relevant for the users. For the ONTO algorithm, we also assessed how using the n most similar items affects the results, with n varying from 1, 5,10,15,20,25,30, and all of the items.…”

“…This work concluded that using semantic similarity improves the classification of the chemical compounds. More recently, [25] used the structural similarity and the ChEBI semantic similarity assembled into a hybrid for predicting compounds subtracts suitable for membrane transporters. Other studies used the semantic similarity of ChEBI entities for recognition and confirmation of chemical compounds found in research documents [26,27].…”