Enhancing Proteostasis Insights: ANN for Ubiquitination Site Prediction

The Ubiquitin-Proteasome System (UPS) is a critical pathway that regulates protein degradation. It is responsible for degrading misfolded, damaged or unwanted protein to maintain cellular homeostasis. Recognizing these complex structures requires knowledge of protein degradation routes and ubiquitination site prediction. In this work, we use machine learning methods to predict protein ubiquitination sites from amino acid sequences. Publicly accessible datasets (HUbipPred_dataset) with annotated ubiquitination sites are used to identify most suitable machine learning algorithm for predicting ubiquitination sites. Machine learning techniques, such as Random Forest (RF) and Artificial Neural Network (ANN) are compared. To enhance prediction accuracy, we incorporate feature selection methods, specifically Minimum Redundancy Maximum Relevance (mRMR) and Elastic Net, to identify the most relevant sequence-based features. From these results, it can be concluded that the ANN model is more effective than other machine learning methods and can be used to predicted ubiquitination site accurately and efficiently.