Incorporating Deep Learning With Word Embedding to Identify Plant Ubiquitylation Sites

Abstract

Protein ubiquitylation is an important posttranslational modification (PTM), which is involved in diverse biological processes and plays an essential role in the regulation of physiological mechanisms and diseases. The Protein Lysine Modifications Database (PLMD) has accumulated abundant ubiquitylated proteins with their substrate sites for more than 20 kinds of species. Numerous works have consequently developed a variety of ubiquitylation site prediction tools across all species, mainly relying on the predefined sequence features and machine learning algorithms. However, the difference in ubiquitylated patterns between these species stays unclear. In this work, the sequence-based characterization of ubiquitylated substrate sites has revealed remarkable differences among plants, animals, and fungi. Then an improved word-embedding scheme based on the transfer learning strategy was incorporated with the multilayer convolutional neural network (CNN) for identifying protein ubiquitylation sites. For the prediction of plant ubiquitylation sites, the proposed deep learning scheme could outperform the machine learning-based methods, with the accuracy of 75.6%, precision of 73.3%, recall of 76.7%, F-score of 0.7493, and 0.82 AUC on the independent testing set. Although the ubiquitylated specificity of substrate sites is complicated, this work has demonstrated that the application of the word-embedding method can enable the extraction of informative features and help the identification of ubiquitylated sites. To accelerate the investigation of protein ubiquitylation, the data sets and source code used in this study are freely available at https://github.com/wang-hong-fei/DL-plant-ubsites-prediction.

Keywords: convolutional neural network; deep learning; plant; transfer learning; ubiquitylation; word embedding.

FIGURE 1

Schematic diagram of the workflow for this study.

FIGURE 2

Word2vec training process of the bigram pattern.

FIGURE 3

Proposed deep structure for ubiquitination site prediction model.

FIGURE 4

Comparison of the amino acid composition (AAC) features between three species. (A) Comparison of AAC features between positive and negative samples of plants. (B) Comparison of AAC features between positive and negative samples of plants. (C) Comparison of AAC features between positive and negative samples of plants.

FIGURE 5

Heatmaps for the amino acid pairwise composition (AAPC) features of three species. (A) Heatmap for the AAPC features of plants. (B) Heatmap for the AAPC features of animals. (C) Heatmap for the AAPC features of fungi.

FIGURE 6

Heatmaps for the positional weighted matrix (PWM) features of three species. (A) Heatmaps for the PWM features of plants. (B) Heatmaps for the PWM features of animals. (C) Heatmaps for the PWM features of fungi.

FIGURE 7

Comparison of the Two Sample Logo of three species. (A) Two Sample Logo of positive samples between plants and animals. (B) Two Sample Logo of positive samples between plants and fungi. (C) Two Sample Logo of positive samples between animals and fungi.

FIGURE 8

ROC curve of the different methods on the testing set.