ABSTRACT
Proteins are one of the essential bio-molecules and control almost all functions in the living cell. They act to perform many of the biological functions in different forms, including catalytic activities in the cell (enzymes), transport proteins (hemoglobin), transmission and coordination between cells and organs of the body (hormones), molecules of the immune system (antibodies), and the transport of small molecules across the cell membranes and so on. Since the origin of life, the protein molecules have attained a specifically designed architecture by nature under the normal physiological conditions. Different functional proteins adopt compact three-dimensional shapes that are biologically active, known as the native conformations. Similarly, if any of the physiological conditions deviate, then the protein extends its conformation. The protein folding problem describes how the native conformation is predicted from a given amino acid sequence. The study of protein folding mechanisms has attracted researchers to conduct intensive research both in experimental and theoretical approaches over the past few years due to its tremendous scope and industrial applications. It is very straightforward to analyze the amino acid sequence of a given protein; however, the prediction of the three-dimensional structure followed by its folding mechanism is a challenging task. For almost 50 years, researchers have been trying to study the protein folding problem using many computational approaches. Many of the artificial intelligence algorithms are suitably used to study the protein folding mechanism. There is a broader scope in applying artificial neural networks (ANN) to study protein folding. As per the recent literature, ANN-based protein folding algorithms are a million times faster than any other algorithms. So in this chapter, the details of ANN-based algorithms and their implementation in the study of protein folding has been discussed by narrating the recent literature.
