Using Honeybees for Gene Expression Profiling: The Artificial Bee Colony Algorithm to Identify Robust Gene Biomarkers for Clinical Diagnosis

One of the popular bioinspired algorithms, the artificial bee colony (ABC), is an evolutionary algorithm based on the foraging behaviour of honey bees for food (nectar). This algorithm works via artificial honeybees which explore several food positions available at a point. Because of the honeybee’s movement, these food positions get altered over time. These honeybees are simply computational agents that help in identifying the best possible solution to a particular problem. Henceforth, ABC algorithm is suitable for clinical data due to its complexity and high dimensionality, making it a combinatorial and continuous optimization problem. Additionally, some hybrid algorithms such as Co-ABC, mRMR-ABC, and SVM-ABC have been developed that work on microarray, RNA-seq, and other high-throughput sequencing-based gene expression profiling allowing for better diagnosis of breast cancers, leukaemia, colon cancers, lung cancer, etc. This chapter gives a peek into the robustness of the ABC and several hybrid ABC algorithms that have been used in gene expression profiling aiding in clinical diagnosis of cancers.