Generic eukaryotic core promoter prediction using structural features of DNA

Abstract

Despite many recent efforts, in silico identification of promoter regions is still in its infancy. However, the accurate identification and delineation of promoter regions is important for several reasons, such as improving genome annotation and devising experiments to study and understand transcriptional regulation. Current methods to identify the core region of promoters require large amounts of high-quality training data and often behave like black box models that output predictions that are difficult to interpret. Here, we present a novel approach for predicting promoters in whole-genome sequences by using large-scale structural properties of DNA. Our technique requires no training, is applicable to many eukaryotic genomes, and performs extremely well in comparison with the best available promoter prediction programs. Moreover, it is fast, simple in design, and has no size constraints, and the results are easily interpretable. We compared our approach with 14 current state-of-the-art implementations using human gene and transcription start site data and analyzed the ENCODE region in more detail. We also validated our method on 12 additional eukaryotic genomes, including vertebrates, invertebrates, plants, fungi, and protists.

Genome-Wide Distribution of TransposedDissociationElements in Maize

“…The various parameters are computationally independent (Liao et al, 2000;Hackett et al, 2007). Such structural computations predict elements of DNA three-dimensional structure, such as local degrees of double helix flexibility versus rigidity (Ornstein et al, 1978;Brukner et al, 1995;Gorin et al, 1995;Olson et al, 1998), and have been used to predict in genomic DNA sequence biologically functional regions, including promoters and transcription start sites (Ohler et al, 2002;Wang and Benham, 2006;Abeel et al, 2008), as well as to profile insertion sites of P element transposons in Drosophila melanogaster and of retroviral and transposon vectors in transgenic animals (Liao et al, 2000;Vigdal et al, 2002;Hackett et al, 2007).…”

“…Finally, different DNA structural preferences for these transposons may also correlate with their different affinities for particular genic regions. For example, promoter regions contain unique DNA structural signatures (Wang and Benham, 2006;Abeel et al, 2008) and are differentially targeted by Mu and Ac/Ds. Overall, it will be useful to exploit a combination of knowledge about local, structural, and generelated preferences with the known regional preference of Ds for hypomethylated regions of the genome to more efficiently design transposon targeting experiments for creating specific gene knockouts.…”

High-resolution human core-promoter prediction with CoreBoost_HM

“…These features may be grouped into two types: one is on small-scale, e.g., TATA-box, GC-box, CAAT-box, and Inr; the other is on larger-scale, such as CpG island, kmer frequency, density of transcription factor binding sites, nucleosome binding, and chromatin modifications (Bajic et al 2004;Zhang 2007). Recently, the large-scale DNA structural features have successfully been used to improve the promoter predictions (Abeel et al 2008). Accordingly, a two-step approach has been proposed for TSS identification (Zhang 1998): First, use the largescale features to roughly locate a promoter in a 1-to 2-kb region (low resolution), then use the small-scale features to refine the prediction into a 100-bp core-promoter region (high resolution).…”

Theoretical and Experimental Investigations of DNA Open States

“…Стабильность ДНК в них зависит от нуклеотидной последовательности исключительно в масштабе единичных пар. В связи с этим профили нестабильности обычно получают путём усреднения по участкам, включающим заданное число пар оснований [270,271].…”

“…Вероятность образования пузырька длиной N пар оснований в большинстве алгоритмов вычислялась путём усреднения [270,271,361]. Однако в 2010 году был разработан подход, позволяющий рассчитывать вероятность синхронного открывания N нуклеотидных пар в модели ближайших соседей [272].…”