Lysine post-translational modifications (PTMs) play a crucial role in regulating diverse functions and biological processes of proteins. However, because of the large volumes of sequencing data generated from genome-sequencing projects, systematic identification of different types of lysine PTM substrates and PTM sites in the entire proteome remains a major challenge. In recent years, a number of computational methods for lysine PTM identification have been developed. These methods show high diversity in their core algorithms, features extracted and feature selection techniques and evaluation strategies. There is therefore an urgent need to revisit these methods and summarize their methodologies, to improve and further develop computational techniques to identify and characterize lysine PTMs from the large amounts of sequence data. With this goal in mind, we first provide a comprehensive survey on a large collection of 49 state-of-the-art approaches for lysine PTM prediction. We cover a variety of important aspects that are crucial for the development of successful predictors, including operating algorithms, sequence and structural features, feature selection, model performance evaluation and software utility. We further provide our thoughts on potential strategies to improve the model performance. Second, in order to examine the feasibility of using deep learning for lysine PTM prediction, we propose a novel computational framework, termed MUscADEL (Multiple Scalable Accurate Deep Learner for lysine PTMs), using deep, bidirectional, long short-term memory recurrent neural networks for accurate and systematic mapping of eight major types of lysine PTMs in the human and mouse proteomes. Extensive benchmarking tests show that MUscADEL outperforms current methods for lysine PTM characterization, demonstrating the potential and power of deep learning techniques in protein PTM prediction. The web server of MUscADEL, together with all the data sets assembled in this study, is freely available at http://muscadel.erc.monash.edu/. We anticipate this comprehensive review and the application of deep learning will provide practical guide and useful insights into PTM prediction and inspire future bioinformatics studies in the related fields.

N⁶-methyladenosine (m⁶A) is a prevalent RNA methylation modification involved in several biological processes. Hundreds or thousands of m⁶A sites identified from different species using high-throughput experiments provides a rich resource to construct in-silico approaches for identifying m⁶A sites. The existing m⁶A predictors are developed using conventional machine-learning (ML) algorithms and most are species-centric. In this paper, we develop a novel cross-species deep-learning classifier based on bidirectional Gated Recurrent Unit (BGRU) for the prediction of m⁶A sites. In comparison with conventional ML approaches, BGRU achieves outstanding performance for the Mammalia dataset that contains over fifty thousand m⁶A sites but inferior for the Saccharomyces cerevisiae dataset that covers around a thousand positives. The accuracy of BGRU is sensitive to the data size and the sensitivity is compensated by the integration of a random forest classifier with a novel encoding of enhanced nucleic acid content. The integrated approach dubbed as BGRU-based Ensemble RNA Methylation site Predictor (BERMP) has competitive performance in both cross-validation test and independent test. BERMP also outperforms existing m⁶A predictors for different species. Therefore, BERMP is a novel multi-species tool for identifying m⁶A sites with high confidence. This classifier is freely available at http://www.bioinfogo.org/bermp.

As a newly-identified protein post-translational modification, malonylation is involved in a variety of biological functions. Recognizing malonylation sites in substrates represents an initial but crucial step in elucidating the molecular mechanisms underlying protein malonylation. In this study, we constructed a deep learning (DL) network classifier based on long short-term memory (LSTM) with word embedding (LSTMWE) for the prediction of mammalian malonylation sites. LSTMWE performs better than traditional classifiers developed with common pre-defined feature encodings or a DL classifier based on LSTM with a one-hot vector. The performance of LSTMWE is sensitive to the size of the training set, but this limitation can be overcome by integration with a traditional machine learning (ML) classifier. Accordingly, an integrated approach called LEMP was developed, which includes LSTMWE and the random forest classifier with a novel encoding of enhanced amino acid content. LEMP performs not only better than the individual classifiers but also superior to the currently-available malonylation predictors. Additionally, it demonstrates a promising performance with a low false positive rate, which is highly useful in the prediction application. Overall, LEMP is a useful tool for easily identifying malonylation sites with high confidence. LEMP is available at http://www.bioinfogo.org/lemp.

Protein lysine crotonylation (Kcr) is an important type of post-translational modification that regulates various activities. The experimental approaches to identify the Kcr sites are time-consuming and it is necessary to develop computational prediction approaches. Previously, a few classifiers were based on over 100 Kcr sites from histone proteins. Recently, thousands of Kcr sites have been experimentally verified on non-histone proteins from the plant species Papaya. We found that the previous classifiers fail to identify non-histone Kcr sites. Therefore, it is necessary to develop classifiers for non-histone proteins. Accordingly, we constructed 11 different classifiers to recognize non-histone Kcr sites by combining different features and algorithms (such as random forest and convolutional neural network (CNN)). They were compared using both ten-fold cross validation and independent test dataset. The classifier based on CNN and the word embedding approach, dubbed as pKcr, performed better than other classifiers. pKcr obtained AUC value of 0.855 and 0.853 for ten-fold cross-validation and independent data test, respectively. No statistical difference of its performances on these two tests indicates that pKcr does not overfit. In the pKcr framework, a peptide is cleaved into biological characters followed by transformation into digital vectors. These vectors are input into the CNN with participation of multiple convolution kernels to automatically extract various features and pooling layers to perform feature selection. The superior performance of pKcr suggests that this algorithm is well suited for the Kcr prediction and may be applied broadly to predicting other types of PTM sites. pKcr can be available at http://www.bioinfogo.org/pkcr.

The unique chemical reactivity of cysteine residues results in various posttranslational modifications (PTMs), which are implicated in regulating a range of fundamental biological processes. With the advent of chemical proteomics technology, thousands of cysteine PTM (CysPTM) sites have been identified from multiple species. A few CysPTM-based databases have been developed, but they mainly focus on data collection rather than various annotations and analytical integration. Here, we present a platform-dubbed CysModDB, integrated with the comprehensive CysPTM resources and analysis tools. CysModDB contains five parts: (1) 70 536 experimentally verified CysPTM sites with annotations of sample origin and enrichment techniques, (2) 21 654 modified proteins annotated with functional regions and structure information, (3) cross-references to external databases such as the protein–protein interactions database, (4) online computational tools for predicting CysPTM sites and (5) integrated analysis tools such as gene enrichment and investigation of sequence features. These parts are integrated using a customized graphic browser and a Basket. The browser uses graphs to represent the distribution of modified sites with different CysPTM types on protein sequences and mapping these sites to the protein structures and functional regions, which assists in exploring cross-talks between the modified sites and their potential effect on protein functions. The Basket connects proteins and CysPTM sites to the analysis tools. In summary, CysModDB is an integrated platform to facilitate the CysPTM research, freely accessible via https://cysmoddb.bioinfogo.org/.

As a novel type of post-translational modification, lysine 2-Hydroxyisobutyrylation (Khib) plays an important role in gene transcription and signal transduction. In order to understand its regulatory mechanism, the essential step is the recognition of Khib sites. Thousands of Khib sites have been experimentally verified across five different species. However, there are only a couple traditional machine-learning algorithms developed to predict Khib sites for limited species, lacking a general prediction algorithm. We constructed a deep-learning algorithm based on convolutional neural network with the one-hot encoding approach, dubbed CNNOH. It performs favorably to the traditional machine-learning models and other deep-learning models across different species, in terms of cross-validation and independent test. The area under the ROC curve (AUC) values for CNNOH ranged from 0.82 to 0.87 for different organisms, which is superior to the currently-available Khib predictors. Moreover, we developed the general model based on the integrated data from multiple species and it showed great universality and effectiveness with the AUC values in the range of 0.79 to 0.87. Accordingly, we constructed the on-line prediction tool dubbed DeepKhib for easily identifying Khib sites, which includes both species-specific and general models. DeepKhib is available at http://www.bioinfogo.org/DeepKhib.