DNA physical properties outperform sequence compositional information in classifying nucleosome-enriched and -depleted regions

Guoqing Liu,Guo-Jun Liu,Jiu-Xin Tan,Hao Lin

DOI: https://doi.org/10.1016/j.ygeno.2018.07.013

IF: 4.31

Genomics

Abstract:The nucleosome is the fundamental structural unit of eukaryotic chromatin and plays an essential role in the epigenetic regulation of cellular processes, such as DNA replication, recombination, and transcription. Hence, it is important to identify nucleosome positions in the genome. Our previous model based on DNA deformation energy, in which a set of DNA physical descriptors was used, performed well in predicting nucleosome dyad positions and occupancy. In this study, we established a machine-learning model for predicting nucleosome occupancy in order to further verify the physical descriptors. Results showed that (1) our model outperformed several other sequence compositional information-based models, indicating a stronger dependence of nucleosome positioning on DNA physical properties; (2) nucleosome-enriched and -depleted regions have distinct features in terms of DNA physical descriptors like sequence-dependent flexibility and equilibrium structure parameters; (3) gene transcription start sites and termination sites can be well characterized with the distribution patterns of the physical descriptors, indicating the regulatory role of DNA physical properties in gene transcription. In addition, we developed a web server for the model, which is freely accessible at http://lin-group.cn/server/iNuc-force/.