Abstract:Glycation, a type of posttranslational modification, preferentially occurs on lysine and arginine residues, impairing protein functionality and altering characteristics. This process is linked to diseases such as Alzheimer's, diabetes, and atherosclerosis. Traditional wet lab experiments are time-consuming, whereas machine learning has significantly streamlined the prediction of protein glycation sites. Despite promising results, challenges remain, including data imbalance, feature redundancy, and suboptimal classifier performance. This research introduces Glypred, a lysine glycation site prediction model combining ClusterCentroids Undersampling (CCU), LightGBM, and bidirectional long short-term memory network (BiLSTM) methodologies, with an additional multihead attention mechanism integrated into the BiLSTM. To achieve this, the study undertakes several key steps: selecting diverse feature types to capture comprehensive protein information, employing a cluster-based undersampling strategy to balance the data set, using LightGBM for feature selection to enhance model performance, and implementing a bidirectional LSTM network for accurate classification. Together, these approaches ensure that Glypred effectively identifies glycation sites with high accuracy and robustness. For feature encoding, five distinct feature types─AAC, KMER, DR, PWAA, and EBGW─were selected to capture a broad spectrum of protein sequence and biological information. These encoded features were integrated and validated to ensure comprehensive protein information acquisition. To address the issue of highly imbalanced positive and negative samples, various undersampling algorithms, including random undersampling, NearMiss, edited nearest neighbor rule, and CCU, were evaluated. CCU was ultimately chosen to remove redundant nonglycated training data, establishing a balanced data set that enhances the model's accuracy and robustness. For feature selection, the LightGBM ensemble learning algorithm was employed to reduce feature dimensionality by identifying the most significant features. This approach accelerates model training, enhances generalization capabilities, and ensures good transferability of the model. Finally, a bidirectional long short-term memory network was used as the classifier, with a network structure designed to capture glycation modification site features from both forward and backward directions. To prevent overfitting, appropriate regularization parameters and dropout rates were introduced, achieving efficient classification. Experimental results show that Glypred achieved optimal performance. This model provides new insights for bioinformatics and encourages the application of similar strategies in other fields. A lysine glycation site prediction software tool was also developed using the PyQt5 library, offering researchers an auxiliary screening tool to reduce workload and improve efficiency. The software and data sets are available on GitHub: https://github.com/ZBYnb/Glypred.

Predicting Golgi-Resident Protein Types Using Conditional Covariance Minimization with XGBoost Based on Multiple Features Fusion

Prediction of Golgi-resident Protein Types by Using Feature Selection Technique

Identify Golgi Protein Types with Modified Mahalanobis Discriminant Algorithm and Pseudo Amino Acid Composition

Prediction of membrane protein types in a hybrid space.

GASIDN: identification of sub-Golgi proteins with multi-scale feature fusion

Identification of sub-Golgi protein localization by use of deep representation learning features

A Random Forest Sub-Golgi Protein Classifier Optimized Via Dipeptide and Amino Acid Composition Features

Predicting protein sub-Golgi locations by combining functional domain enrichment scores with pseudo-amino acid compositions.

Golgi_DF: Golgi proteins classification with deep forest

Protein Function Prediction With Functional and Topological Knowledge of Gene Ontology

A Novel Protein Subcellular Localization Method with CNN-XGBoost Model for Alzheimer's Disease.

Glypre: In Silico Prediction Of Protein Glycation Sites By Fusing Multiple Features And Support Vector Machine

A New Hybrid Approach to Predict Subcellular Localization by Incorporating Protein Evolutionary Conservation Information

Prediction of Protein Subcellular Localization Based on Fusion of Multi-view Features

Developing Computational Model to Predict Protein-Protein Interaction Sites Based on the XGBoost Algorithm

Glypred: Lysine Glycation Site Prediction via CCU-LightGBM-BiLSTM Framework with Multi-Head Attention Mechanism

Prediction for Membrane Protein Types Based on Effective Fusion Representation and MIC-GA Feature Selection

Improving protein-protein interactions prediction accuracy using XGBoost feature selection and stacked ensemble classifier

Globally predicting protein functions based on co-expressed protein-protein interaction networks and ontology taxonomy similarities.

Predicting G-protein coupled receptors-G-protein coupling specificity based on autocross-covariance transform.

Predicting Gram-Positive Bacterial Protein Subcellular Location by Using Combined Features