Implementing Deep Learning Algorithm on Physicochemical Properties of Proteins

In recent years, the increase of complex protein data reveals the emergence of deep learning in the data mining field. The number of research has shown it as one of the powerful tools that transform these big data to valuable information or knowledge. The prediction of structure of proteins contributes to its functionality which can be used for drug discovery, medicine design and other important areas. The amino acids physicochemical properties determine the protein structure quality, which can further identify the difference between native and predicted proteins. In this paper, the dataset considered has nine physicochemical properties, and they are used to determine the root mean square deviation (RMSD). A deep learning model is applied to efficiently implement the model, and the performance of the model is evaluated on the basis of root mean-squared error (RMSE) and the value of R-squared (R2) which is 3.71 and 0.6327, respectively.