Predicting Chromosome Flexibility from the Genomic Sequence Based on Deep Learning Neural Networks

Background: The open and accessible regions of the chromosome are more likely to be bound by transcription factors which are important for nuclear processes and biological functions. Studying the change of chromosome flexibility can help to discover and analyze disease markers and improve the efficiency of clinical diagnosis. Current methods for predicting chromosome flexibility based on Hi-C data include the flexibility-rigidity index (FRI) and the Gaussian network model (GNM), which have been proposed to characterize chromosome flexibility. However, these methods require the chromosome structure data based on 3D biological experiments, which is time-consuming and expensive. Objective: Generally, the folding and curling of the double helix sequence of DNA have a great impact on chromosome flexibility and function. Motivated by the success of genomic sequence analysis in biomolecular function analysis, we hope to propose a method to predict chromosome flexibility only based on genomic sequence data. Method: We propose a new method (named "DeepCFP") using deep learning models to predict chromosome flexibility based on only genomic sequence features. The model has been tested in the GM12878 cell line. Results: The maximum accuracy of our model has reached 91%. The performance of DeepCFP is close to FRI and GNM. Conclusion: The DeepCFP can achieve high performance only based on genomic sequence.

DiNeR: a Differential Graphical Model for analysis of co-regulation Network Rewiring

BackgroundDuring transcription, numerous transcription factors (TFs) bind to targets in a highly coordinated manner to control the gene expression. Alterations in groups of TF-binding profiles (i.e. “co-binding changes”) can affect the co-regulating associations between TFs (i.e. “rewiring the co-regulator network”). This, in turn, can potentially drive downstream expression changes, phenotypic variation, and even disease. However, quantification of co-regulatory network rewiring has not been comprehensively studied.MethodsTo address this, we propose DiNeR, a computational method to directly construct a differential TF co-regulation network from paired disease-to-normal ChIP-seq data. Specifically, DiNeR uses a graphical model to capture the gained and lost edges in the co-regulation network. Then, it adopts a stability-based, sparsity-tuning criterion -- by sub-sampling the complete binding profiles to remove spurious edges -- to report only significant co-regulation alterations. Finally, DiNeR highlights hubs in the resultant differential network as key TFs associated with disease.ResultsWe assembled genome-wide binding profiles of 104 TFs in the K562 and GM12878 cell lines, which loosely model the transition between normal and cancerous states in chronic myeloid leukemia (CML). In total, we identified 351 significantly altered TF co-regulation pairs. In particular, we found that the co-binding of the tumor suppressor BRCA1 and RNA polymerase II, a well-known transcriptional pair in healthy cells, was disrupted in tumors. Thus, DiNeR successfully extracted hub regulators and discovered well-known risk genes.ConclusionsOur method DiNeR makes it possible to quantify changes in co-regulatory networks and identify alterations to TF co-binding patterns, highlighting key disease regulators. Our method DiNeR makes it possible to quantify changes in co-regulatory networks and identify alterations to TF co-binding patterns, highlighting key disease regulators.

Long-TUC-seq is a robust method for quantification of metabolically labeled full-length isoforms

ABSTRACTThe steady state expression of each gene is the result of a dynamic transcription and degradation of that gene. While regular RNA-seq methods only measure steady state expression levels, RNA-seq of metabolically labeled RNA identifies transcripts that were transcribed during the window of metabolic labeling. Whereas short-read RNA sequencing can identify metabolically labeled RNA at the gene level, long-read sequencing provides much better resolution of isoform-level transcription. Here we combine thiouridine-to-cytosine conversion (TUC) with PacBio long-read sequencing to study the dynamics of mRNA transcription in the GM12878 cell line. We show that using long-TUC-seq, we can detect metabolically labeled mRNA of distinct isoforms more reliably than using short reads. Long-TUC-seq holds the promise of capturing isoform dynamics robustly and without the need for enrichment.

Revealing Hi-C subcompartments by imputing inter-chromosomal chromatin interactions

Higher-order genome organization and its variation in different cellular conditions remain poorly understood. Recent high-coverage genome-wide chromatin interaction mapping using Hi-C has revealed spatial segregation of chromosomes in the human genome into distinct subcompartments. However, subcompartment annotation, which requires Hi-C data with high sequencing coverage, is currently only available in the GM12878 cell line, making it impractical to compare subcompartment patterns across cell types. Here we develop a computational approach, SNIPER (Subcompartment iNference using Imputed Probabilistic ExpRessions), based on denoising autoencoder and multilayer perceptron classifier to infer subcompartments using typical Hi-C datasets with moderate coverage. SNIPER accurately reveals subcompartments using moderate coverage Hi-C datasets and outperforms an existing method that uses epigenomic features in GM12878. We apply SNIPER to eight additional cell lines and find that chromosomal regions with conserved and cell-type specific subcompartment annotations have different patterns of functional genomic features. SNIPER enables the identification of subcompartments without high-coverage Hi-C data and provides insights into the function and mechanisms of spatial genome organization variation across cell types.

A technology-agnostic long-read analysis pipeline for transcriptome discovery and quantification

ABSTRACTAlternative splicing is widely acknowledged to be a crucial regulator of gene expression and is a key contributor to both normal developmental processes and disease states. While cost-effective and accurate for quantification, short-read RNA-seq lacks the ability to resolve full-length transcript isoforms despite increasingly sophisticated computational methods. Long-read sequencing platforms such as Pacific Biosciences (PacBio) and Oxford Nanopore (ONT) bypass the transcript reconstruction challenges of short reads. Here we introduce TALON, the ENCODE4 pipeline for platform-independent analysis of long-read transcriptomes. We apply TALON to the GM12878 cell line and show that while both PacBio and ONT technologies perform well at full-transcript discovery and quantification, each displayed distinct technical artifacts. We further apply TALON to mouse hippocampus and cortex transcriptomes and find that 422 genes found in these regions have more reads associated with novel isoforms than with annotated ones. We demonstrate that TALON is a capable of tracking both known and novel transcript models as well as their expression levels across datasets for both simple studies and in larger projects. These properties will enable TALON users to move beyond the limitations of short-read data to perform isoform discovery and quantification in a uniform manner on existing and future long-read platforms.

Revealing Hi-C subcompartments by imputing high-resolution inter-chromosomal chromatin interactions

The higher-order genome organization and its variation in different cellular conditions remains poorly understood. Recent high-resolution genome-wide mapping of chromatin interactions using Hi-C has revealed that chromosomes in the human genome are spatially segregated into distinct subcompartments. However, due to the requirement on sequencing coverage of the Hi-C data to define subcompartments, to date subcompartment annotation is only available in the GM12878 cell line, making it impractical to compare Hi-C subcompartment patterns across multiple cell types. Here we develop a new computational approach, named Sniper, based on an autoencoder and multilayer perceptron classifier to infer subcompartments using typical Hi-C datasets with moderate coverage. We demonstrated that Sniper can accurately reveal subcompartments based on Hi-C datasets with moderate coverage and can significantly outperform an existing method that uses numerous epigenomic datasets as input features in GM12878. We applied Sniper to eight additional cell lines to identify the variation of Hi-C subcompartments across different cell types. Sniper revealed that chromosomal regions with conserved and more dynamic subcompartment annotations across cell types have different patterns of functional genomic features. This work demonstrates that Sniper is effective in identifying subcompartments without the need of high-coverage Hi-C data and has the potential to provide new insights into the spatial genome organization variation across different cell types.

Nucleotide sequence and DNaseI sensitivity are predictive of 3D chromatin architecture

Recently, Hi-C has been used to probe the 3D chromatin architecture of multiple organisms and cell types. The resulting collections of pairwise contacts across the genome have connected chromatin architecture to many cellular phenomena, including replication timing and gene regulation. However, high resolution (10 kb or finer) contact maps remain scarce due to the expense and time required for collection. A computational method for predicting pairwise contacts without the need to run a Hi-C experiment would be invaluable in understanding the role that 3D chromatin architecture plays in genome biology. We describe Rambutan, a deep convolutional neural network that predicts Hi-C contacts at 1 kb resolution using nucleotide sequence and DNaseI assay signal as inputs. Specifically, Rambutan identifies locus pairs that engage in high confidence contacts according to Fit-Hi-C, a previously described method for assigning statistical confidence estimates to Hi-C contacts. We first demonstrate Rambutan’s performance across chromosomes at 1 kb resolution in the GM12878 cell line. Subsequently, we measure Rambutan’s performance across six cell types. In this setting, the model achieves an area under the receiver operating characteristic curve between 0.7662 and 0.8246 and an area under the precision-recall curve between 0.3737 and 0.9008. We further demonstrate that the predicted contacts exhibit expected trends relative to histone modification ChlP-seq data, replication timing measurements, and annotations of functional elements such as promoters and enhancers. Finally, we predict Hi-C contacts for 53 human cell types and show that the predictions cluster by cellular function. [NOTE: After our original submission we discovered an error in our calling of statistically significant contacts. Briefly, when calculating the prior probability of a contact, we used the number of contacts at a certain genomic distance in a chromosome but divided by the total number of bins in the full genome. When we corrected this mistake we noticed that the Rambutan model, as it curently stands, did not outperform simply using the GM12878 contact map that Rambutan was trained on as the predictor in other cell types. While we investigate these new results, we ask that readers treat this manuscript skeptically.]

A semi-supervised approach uncovers thousands of intragenic enhancers differentially activated in human cells

Background Transcriptional enhancers are generally known to regulate gene transcription from afar. Their activation involves a series of changes in chromatin marks and recruitment of protein factors. These enhancers may also occur inside genes, but how many may be active in human cells and their effects on the regulation of the host gene remains unclear. Results We describe a novel semi-supervised method based on the relative enrichment of chromatin signals between 2 conditions to predict active enhancers. We applied this method to the tumoral K562 and the normal GM12878 cell lines to predict enhancers that are differentially active in one cell type. These predictions show enhancer-like properties according to positional distribution, correlation with gene expression and production of enhancer RNAs. Using this model, we predict 10,365 and 9,777 intragenic active enhancers in K562 and GM12878, respectively, and relate the differential activation of these enhancers to expression and splicing differences of the host genes. Conclusions We propose that the activation or silencing of intragenic transcriptional enhancers modulate the regulation of the host gene by means of a local change of the chromatin and the recruitment of enhancer-related factors that may interact with the RNA directly or through the interaction with RNA binding proteins. Predicted enhancers are available at http://regulatorygenomics.upf.edu/Projects/enhancers.html