Artificial Immune Network Algorithm for the Multiple Sequence Alignment Problem of Alzheimer’s Disease Amyloid-Secretase-Pathway

This study investigates the use of the aiNet bioinspired metaheuristic algorithm in the multiple sequence alignment problem (MSA), a core problem in bioinformatics. We specifically focus on the alignment of genetic sequences from Alzheimer’s Disease Amyloid-Secretase-Pathway, a critical disease with prevalence in the older population. The aiNet algorithm is based on the human autoimmune system theory, where specialized cells evolve to adapt their characteristics to eliminate invader pathogens in the organism. This unique optimization mechanism inspires the fundamentals of our algorithm. Our work involves three sequential adjustment phases for the considered parameters: population size, Clon Rate,elimination Threshold, number of Cycles,number of generated Clones, and the Random Clone Insertion value. These phases are designed to fine-tune the algorithm and improve its performance. To evaluate the results, we compare three metrics of interest: The achieved Fitness values of the solutions found, the number of functions being assessed required to accomplish the process (NFE), and the time needed for the algorithm to process the alignment problem. The results demonstrate the promising potential of the aiNet algorithm in optimization and bioinformatics, particularly in improving fitness values. However, it’s important to note that hardware resources play a significant role when processing biological data. Additionally, the time and NFE values may be influenced by the cell population size and the number of optimization cycles.