Computational discovery of feature patterns in nucleosomal DNA sequences
Abbreviated Journal Title
Nucleosome formation; Feature; Pattern discovery; HIGH-RESOLUTION; HUMAN GENOME; SACCHAROMYCES-CEREVISIAE; STRUCTURAL; FEATURES; TRANSCRIPTION; OCCUPANCY; POSITIONS; CHROMATIN; PROMOTER; YEAST; Biotechnology & Applied Microbiology; Genetics & Heredity
The identification of important factors that affect nucleosome formation is critical to clarify nucleosome-forming mechanisms and the role of the nucleosome in gene regulation. Various features reported in the literature led to our hypothesis that multiple features can together contribute to nucleosome formation. Therefore, we compiled 779 features and developed a pattern discovery and scoring algorithm FFN (Finding Features for Nucleosomes) to identify feature patterns that are differentially enriched in nucleosome-forming sequences and nucleosome-depletion sequences. Applying FFN to genome-wide nucleosome occupancy data in yeast and human, we identified statistically significant feature patterns that may influence nucleosome formation, many of which are common to the two species. We found that both sequence and structural features are important in nucleosome occupancy prediction. We discovered that, even for the same feature combinations, variations in feature values may lead to differences in predictive power. We demonstrated that the identified feature patterns could be used to assist nucleosomal sequence prediction. (C) 2014 Elsevier Inc. All rights reserved.
"Computational discovery of feature patterns in nucleosomal DNA sequences" (2014). Faculty Bibliography 2010s. 6369.