PScL-DDCFPred: an ensemble deep learning-based approach for characterizing multiclass subcellular localization of human proteins from bioimage data

Abstract

Motivation: Characterization of protein subcellular localization has become an important and long-standing task in bioinformatics and computational biology, which provides valuable information for elucidating various cellular functions of proteins and guiding drug design.

Results: Here, we develop a novel bioimage-based computational approach, termed PScL-DDCFPred, to accurately predict protein subcellular localizations in human tissues. PScL-DDCFPred first extracts multiview image features, including global and local features, as base or pure features; next, it applies a new integrative feature selection method based on stepwise discriminant analysis and generalized discriminant analysis to identify the optimal feature sets from the extracted pure features; Finally, a classifier based on deep neural network (DNN) and deep-cascade forest (DCF) is established. Stringent 10-fold cross-validation tests on the new protein subcellular localization training dataset, constructed from the human protein atlas databank, illustrates that PScL-DDCFPred achieves a better performance than several existing state-of-the-art methods. Moreover, the independent test set further illustrates the generalization capability and superiority of PScL-DDCFPred over existing predictors. In-depth analysis shows that the excellent performance of PScL-DDCFPred can be attributed to three critical factors, namely the effective combination of the DNN and DCF models, complementarity of global and local features, and use of the optimal feature sets selected by the integrative feature selection algorithm.

Availability and implementation: https://github.com/csbio-njust-edu/PScL-DDCFPred.

Supplementary information: Supplementary data are available at Bioinformatics online.

Fig. 1.

The workflow of the proposed PScL-DDCFPred approach

Fig. 2.

Performance comparison of the models trained using the pure feature sets and optimal feature sets obtained after the SDA-GDA feature selection. Panels (A), (B) and (C) show the performance comparison in terms of OA, F1-Score_M and MCC, respectively

Fig. 3.

Performance comparison of models trained using the RICLBP pure and optimal feature sets. Panels (A) and (B) show the PR Curves while (C) and (D) show the ROC curves for the RICLBP pure and optimal feature sets, respectively

Fig. 4.

Performance comparison of the classical DCF, DNN and proposed DNN-DCF classifiers. Panels (A), (B), (C) and (D) show the performance comparison for all the three classifiers in terms of meanAUC, meanAUPR, stdAUC and stdAUPR, respectively

Fig. 5.

Performance comparison of PScL-DDCFPred and the recently published PSc-HDeep method. Panels (A) and (B) show the PR Curves while (C) and (D) show the ROC curves of PSc-HDeep and PScL-DDCFPred, respectively

Fig. 6.

Performance comparison of the proposed PScL-DDCFPred and the existing state-of-the-art methods on the independent test dataset