Published Paper Details:

Amino acids are traditionally categorized based on their biochemical attributes. In this study, the focus shifts to reorganizing amino acids solely according to structural statistics, thereby mitigating existing chemical biases. The proposed machine learning model is to propose classification of the novel amino acid based on their structural insights. Main aim is to study the protein sequences of the candidates,that have been crystallized and their X-ray structures are well known and deposited in Protein Data Bank (PDB). Growth in available amino acids eventually slows down the speed of identifying the protein constituted by the number of amino acids. Hence, there is a requirement for an effective model to predict the secondary protein structure swiftly and accurately. This model has been designed to anticipate the protein's secondary structure through specific predefined tasks. Our Aim is to study the protein sequences of the candidates who have been crystallized and their X-ray structures are known and deposited in Research Collaboratory for Structural Bioinformatics (RCSB) PDB. The sequences of all the known crystal structures are to be downloaded from PDB and then a table is to be prepared based on occurrence of each amino acid on independent positions in the range of secondary structure. Once the data is ready, the noise must be removed accordingly, and the inferences will be made by polishing the data above a reasonable threshold. This focuses on forecasting the required parameters by analyzing the arrangement of amino acids and their neighboring context.