CpG islands are generally known as the epigenetic regulatory regions in accordance with histone modifications, methylation,and promoter activity. There is a significant need for the exact mapping of DNA methylation in CpG islands tounderstand the diverse biological functions. However, the precise identification of CpG islands from the whole genomethrough experimental and computational approaches is still challenging. Numerous computational methods are beingdeveloped to detect the CpG-enriched regions, effectively, to reduce the time and cost of the experiments. Here, we reviewsome of the latest computational CpG detection methods that utilize clustering, patterns and physical-distance likeparameters for CpG island detection. The comparative analyses of the methods relying on different principles andparameters allow prioritizing the algorithms for specific CpG associated datasets to achieve higher accuracy and sensitivity.A number of computational tools based on the window, Hidden Markov Model, density and distance-/length-basedalgorithms are being applied on human or mammalian genomes for accurate CpG detection. Comparative analyses of CpGisland detection algorithms facilitate to prefer the method according to the target genome and required parameters to attainhigher accuracy, specificity, and performance. There is still a need for efficient computational CpG detection methods withlower false-positive results. This review provides a better understanding about the principles of tools that will assist toprioritize and develop the algorithms for accurate CpG islands detection.
Volume 45, 2020
Continuous Article Publishing mode
Click here for Editorial Note on CAP Mode