Un modelo explicable de aprendizaje profundo para la predicción de proteínas de clatrina utilizando una matriz de...

Clathrin protein (CP) plays a vital role in essential cellular processes, in-cluding endocytosis and signal transduction. Dysfunctions in CP have been associated with vari-ous diseases, including neurodegenerative disorders and cancer, making it a significant focus of biomedical research.

This study introduces Pred-CLGRUs, a novel computational framework for predicting Clathrin protein using Gated Recurrent Units (GRUs). Pred-CLGRUs utilize a Position-Specific Scoring Matrix (PSSM) in conjunction with the Discrete Cosine Transform (DCT) to enhance feature extraction while reducing noise and redundancy. A potential explanation for this mis-match could be that the Average Block-PSSM feature is derived from the Position-Specific Scor-ing Matrix (PSSM). The proposed model is trained using four deep learning architectures: GRUs, Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, and Deep Neural Networks (DNNs).

The PSSM-DCT input enabled GRUs to achieve 93.33% accuracy, producing a Mat-thews Correlation Coefficient value of 0.86. Pred-CLGRUs enabled GRUs to achieve an accuracy of 93.33%, sensitivity of 94.44%, and specificity of 92.22%, while maintaining a Matthews corre-lation coefficient of 0.86.

We began by calculating PSSM to detect evolutionary changes. The compression method named DCT was applied within each PSSM during the second implementation step.

The predictive performance can be significantly enhanced if the model can incorpo-rate heterogeneous types of biological information, such as structure, evolution, and function. We plan to design a novel prediction model that combines Capsule Neural Networks (CapsNet) with DeepWalk to obtain structural features.