CNNAthUbi

CNNAthUbi predicts ubiquitination sites in the Arabidopsis thaliana proteome using convolutional neural network (CNN) models trained on sequence-derived features.

Key Features:

• CNN-Based Ubiquitination Site Prediction: Implements two convolutional neural network models to identify lysine ubiquitination sites in Arabidopsis thaliana proteins.
• Sequence Feature Analysis: Incorporates physicochemical properties of amino acids within sequence regions surrounding ubiquitination sites to improve predictive performance.
• Model Architecture Evaluation: Examines the impact of CNN structural configurations on prediction accuracy.
• Proteome-Wide Prediction: Generates predicted ubiquitination sites across the global Arabidopsis thaliana proteome.

Scientific Applications:

• Post-Translational Modification Analysis: Identifies candidate ubiquitination sites to investigate protein regulation in Arabidopsis thaliana.
• Plant Functional Genomics: Supports studies of ubiquitination-mediated processes including plant development, metabolism, and stress responses.

Methodology:

CNNAthUbi applies convolutional neural network models trained on sequence-derived physicochemical features surrounding lysine residues and evaluates performance using five-fold cross-validation and independent test datasets.