Dataset Information

Combining Hi-C data with phylogenetic correlation to predict the target genes of distal regulatory elements in human genome.

ABSTRACT: Defining the target genes of distal regulatory elements (DREs), such as enhancer, repressors and insulators, is a challenging task. The recently developed Hi-C technology is designed to capture chromosome conformation structure by high-throughput sequencing, and can be potentially used to determine the target genes of DREs. However, Hi-C data are noisy, making it difficult to directly use Hi-C data to identify DRE-target gene relationships. In this study, we show that DREs-gene pairs that are confirmed by Hi-C data are strongly phylogenetic correlated, and have thus developed a method that combines Hi-C read counts with phylogenetic correlation to predict long-range DRE-target gene relationships. Analysis of predicted DRE-target gene pairs shows that genes regulated by large number of DREs tend to have essential functions, and genes regulated by the same DREs tend to be functionally related and co-expressed. In addition, we show with a couple of examples that the predicted target genes of DREs can help explain the causal roles of disease-associated single-nucleotide polymorphisms located in the DREs. As such, these predictions will be of importance not only for our understanding of the function of DREs but also for elucidating the causal roles of disease-associated noncoding single-nucleotide polymorphisms.

SUBMITTER: Lu Y

PROVIDER: S-EPMC3905853 | biostudies-literature | 2013 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Combining Hi-C data with phylogenetic correlation to predict the target genes of distal regulatory elements in human genome.

Lu Yulan Y Zhou Yuanpeng Y Tian Weidong W

Nucleic acids research 20130903 22

Defining the target genes of distal regulatory elements (DREs), such as enhancer, repressors and insulators, is a challenging task. The recently developed Hi-C technology is designed to capture chromosome conformation structure by high-throughput sequencing, and can be potentially used to determine the target genes of DREs. However, Hi-C data are noisy, making it difficult to directly use Hi-C data to identify DRE-target gene relationships. In this study, we show that DREs-gene pairs that are co ...[more]

PMID: 24003029

Similar Datasets

Project description:BackgroundThe data deluge can leverage sophisticated ML techniques for functionally annotating the regulatory non-coding genome. The challenge lies in selecting the appropriate classifier for the specific functional annotation problem, within the bounds of the hardware constraints and the model's complexity. In our system AIKYATAN, we annotate distal epigenomic regulatory sites, e.g., enhancers. Specifically, we develop a binary classifier that classifies genome sequences as distal regulatory regions or not, given their histone modifications' combinatorial signatures. This problem is challenging because the regulatory regions are distal to the genes, with diverse signatures across classes (e.g., enhancers and insulators) and even within each class (e.g., different enhancer sub-classes).ResultsWe develop a suite of ML models, under the banner AIKYATAN, including SVM models, random forest variants, and deep learning architectures, for distal regulatory element (DRE) detection. We demonstrate, with strong empirical evidence, deep learning approaches have a computational advantage. Plus, convolutional neural networks (CNN) provide the best-in-class accuracy, superior to the vanilla variant. With the human embryonic cell line H1, CNN achieves an accuracy of 97.9% and an order of magnitude lower runtime than the kernel SVM. Running on a GPU, the training time is sped up 21x and 30x (over CPU) for DNN and CNN, respectively. Finally, our CNN model enjoys superior prediction performance vis-'a-vis the competition. Specifically, AIKYATAN-CNN achieved 40% higher validation rate versus CSIANN and the same accuracy as RFECS.ConclusionsOur exhaustive experiments using an array of ML tools validate the need for a model that is not only expressive but can scale with increasing data volumes and diversity. In addition, a subset of these datasets have image-like properties and benefit from spatial pooling of features. Our AIKYATAN suite leverages diverse epigenomic datasets that can then be modeled using CNNs with optimized activation and pooling functions. The goal is to capture the salient features of the integrated epigenomic datasets for deciphering the distal (non-coding) regulatory elements, which have been found to be associated with functional variants. Our source code will be made publicly available at: https://bitbucket.org/cellsandmachines/aikyatan.

Project description:The Lim domain only 2 (Lmo2) gene encodes a transcriptional cofactor critical for the development of hematopoietic stem cells. Several distal regulatory elements have been identified upstream of the Lmo2 gene in the human and mouse genomes that are capable of enhancing reporter gene expression in erythroid cells and may be responsible for the high level transcription of Lmo2 in the erythroid lineage. In this study we investigate how these elements regulate transcription of Lmo2 and whether or not they function cooperatively in the endogenous context. Chromosome conformation capture (3C) experiments show that chromatin-chromatin interactions exist between upstream regulatory elements and the Lmo2 promoter in erythroid cells but that these interactions are absent from kidney where Lmo2 is transcribed at twelve fold lower levels. Specifically, long range chromatin-chromatin interactions occur between the Lmo2 proximal promoter and two broad regions, 3-31 and 66-105 kb upstream of Lmo2, which we term the proximal and distal control regions for Lmo2 (pCR and dCR respectively). Each of these regions is bound by several transcription factors suggesting that multiple regulatory elements cooperate in regulating high level transcription of Lmo2 in erythroid cells. Binding of CTCF and cohesin which support chromatin loops at other loci were also found within the dCR and at the Lmo2 proximal promoter. Intergenic transcription occurs throughout the dCR in erythroid cells but not in kidney suggesting a role for these intergenic transcripts in regulating Lmo2, similar to the broad domain of intergenic transcription observed at the human β-globin locus control region. Our data supports a model in which the dCR functions through a chromatin looping mechanism to contact and enhance Lmo2 transcription specifically in erythroid cells. Furthermore, these chromatin loops are supported by the cohesin complex recruited to both CTCF and transcription factor bound regions.

Dataset Information

Combining Hi-C data with phylogenetic correlation to predict the target genes of distal regulatory elements in human genome.

Publications

Combining Hi-C data with phylogenetic correlation to predict the target genes of distal regulatory elements in human genome.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets