Computational Approaches to Predict the Non-canonical DNAs

TLDR: This study reviews the currently available computational methods for predicting the non-canonical DNAs in the genome and reviews strategies for the identification of ncDNA motifs across the whole genome, necessary for further understanding and investigation of the structure and function of nCDNAs.

Abstract: \n\nAlthough most nucleotides in the genome form canonical double-stranded\nB-DNA, many repeated sequences transiently present as non-canonical conformations (non-B\nDNA) such as triplexes, quadruplexes, Z-DNA, cruciforms, and slipped/hairpins. Those noncanonical\nDNAs (ncDNAs) are not only associated with many genetic events such as replication,\ntranscription, and recombination, but are also related to the genetic instability that results in the\npredisposition to disease. Due to the crucial roles of ncDNAs in cellular and genetic functions,\nvarious computational methods have been implemented to predict sequence motifs that generate\nncDNA.\n\n\n\nHere, we review strategies for the identification of ncDNA motifs across the whole\ngenome, which is necessary for further understanding and investigation of the structure and\nfunction of ncDNAs.\n\n\n\nThere is a great demand for computational prediction of non-canonical DNAs that\nplay key functional roles in gene expression and genome biology. In this study, we review the\ncurrently available computational methods for predicting the non-canonical DNAs in the genome.\nCurrent studies not only provide an insight into the computational methods for predicting the\nsecondary structures of DNA but also increase our understanding of the roles of non-canonical\nDNA in the genome.\n