HIV-Prevalence-Prediction Summary:

• This project investigates the progression of the human immunodeficiency virus (HIV) by finding markers in the HIV sequence that predict the patient's response to the treatment over 16 weeks.
• Four main components have been used to predict the severity of the infection: sequences of patients' reverse transcriptase (RT) and their protease (PR), the viral load, and the CD4 count.
• The main purpose of this study is to examine the use of different encoding techniques for both RT and PR sequences.
• The Back Propagation-learning algorithm has been utilized to train the input data for a number of patients using multilayer perceptron (MLP) neural network.
• A variety of encoding schemes for genetic sequences were explored and utilized: Enumeration (80.43% Accuracy), Five Bit Encoding (78.62% Accuracy), Orthogonal Encoding (63.21% Accuracy), N-grams (73.21% Accuracy), Enzyme Properties (67.39% Accuracy).