scHiCTools: A computational toolbox for analyzing single-cell Hi-C data

Single-cell Hi-C (scHi-C) sequencing technologies allow us to investigate three-dimensional chromatin organization at the single-cell level. However, we still need computational tools to deal with the sparsity of the contact maps from single cells and embed single cells in a lower-dimensional Euclidean space. This embedding helps us understand relationships between the cells in different dimensions, such as cell-cycle dynamics and cell differentiation. We present an open-source computational toolbox, scHiCTools, for analyzing single-cell Hi-C data comprehensively and efficiently. The toolbox provides two methods for screening single cells, three common methods for smoothing scHi-C data, three efficient methods for calculating the pairwise similarity of cells, three methods for embedding single cells, three methods for clustering cells, and a build-in function to visualize the cells embedding in a two-dimensional or three-dimensional plot. scHiCTools, written in Python3, is compatible with different platforms, including Linux, macOS, and Windows.     

Clustering Deviation Index (CDI): a robust and accurate internal measure for evaluating scRNA-seq data clustering

Single-cell high-throughput chromatin conformation capture methodologies (scHi-C) enable profiling of long-range genomic interactions. However, data from these technologies are prone to technical noise and biases that hinder downstream analysis. We develop a normalization approach, BandNorm, and a deep generative modeling framework, scVI-3D, to account for scHi-C specific biases. In benchmarking experiments, BandNorm yields leading performances in a time and memory efficient manner for cell-type separation, identification of interacting loci, and recovery of cell-type relationships, while scVI-3D exhibits advantages for rare cell types and under high sparsity scenarios. Application of BandNorm coupled with gene-associating domain analysis reveals scRNA-seq validated sub-cell type identification.     

THUNDER: A reference-free deconvolution method to infer cell type proportions from bulk Hi-C data

Hi-C data provide population averaged estimates of three-dimensional chromatin contacts across cell types and states in bulk samples. Effective analysis of Hi-C data entails controlling for the potential confounding factor of differential cell type proportions across heterogeneous bulk samples. We propose a novel unsupervised deconvolution method for inferring cell type composition from bulk Hi-C data, the Two-step Hi-c UNsupervised DEconvolution appRoach (THUNDER). We conducted extensive simulations to test THUNDER based on combining two published single-cell Hi-C (scHi-C) datasets. THUNDER more accurately estimates the underlying cell type proportions compared to reference-free methods (e.g., TOAST, and NMF) and is more robust than reference-dependent methods (e.g. MuSiC). We further demonstrate the practical utility of THUNDER to estimate cell type proportions and identify cell-type-specific interactions in Hi-C data from adult human cortex tissue samples. THUNDER will be a useful tool in adjusting for varying cell type composition in population samples, facilitating valid and more powerful downstream analysis such as differential chromatin organization studies. Additionally, THUNDER estimated contact profiles provide a useful exploratory framework to investigate cell-type-specificity of the chromatin interactome while experimental data is still rare.     

Translocation detection from Hi-C data via scan statistics

Recent Hi-C technology enables more comprehensive chromosomal conformation research, including the detection of structural variations, especially translocations. In this paper, we formulate the interchromosomal translocation detection as a problem of scan clustering in a spatial point process. We then develop TranScan, a new translocation detection method through scan statistics with the control of false discovery. The simulation shows that TranScan is more powerful than an existing sophisticated scan clustering method, especially under strong signal situations. Evaluation of TranScan against current translocation detection methods on realistic breakpoint simulations generated from real data suggests better discriminative power under the receiver-operating characteristic curve. Power analysis also highlights TranScan's consistent outperformance when sequencing depth and heterozygosity rate is varied. Comparatively, Type I error rate is lowest when evaluated using a karyotypically normal cell line. Both the simulation and real data analysis indicate that TranScan has great potentials in interchromosomal translocation detection using Hi-C data.     

A novel method to identify topological domains using Hi-C data

Over the last decade the 3C-based (Chromosome Conformation Capture, 3C) approaches have been developed to describe the frequency of chromatin interaction. The invention of Hi-C allows us to obtain genome-wide chromatin interaction map. However, it is challenging to develop efficient and robust analytical tools to interpret the Hi-C data. Here we present a new method called Clustering based Hi-C Domain Finder (CHDF), which is based on the difference of interaction intensity inside/outside domains, to identify Hi-C domains. We also compared CHDF with existing methods including Direction Index (DI) and HiCseg. CHDF can define more chromatin domains validated by higher resolution local chromatin structure data (Chromosome Conformation Capture Carbon Copy (5C) data). Using Hi-C data of lower sequencing depth, chromatin structure identified by CHDF is closer to that discovered by data of higher sequencing depth. Furthermore, the implement of CHDF is faster than the other two. Using CHDF, we are potentially able to discover more hints and clues about chromatin structural elements at domain level.     

Combined analyses of RNA-sequence and Hi-C along with GWAS loci—A novel approach to dissect keloid disorder genetic mechanism

Keloid disorder is a tumour-like disease with invasive growth and a high recurrence rate. Genetic contribution is well expected due to the presence of autosomal dominant inheritance and various genetic mutations in keloid lesions. However, GWAS failed to reveal functional variants in exon regions but single nucleotide polymorphisms in the non-coding regions, suggesting the necessity of innovative genetic investigation. This study employed combined GWAS, RNA-sequence and Hi-C analyses to dissect keloid disorder genetic mechanisms using paired keloid tissues and normal skins. Differentially expressed genes, miRNAs and lncRNAs mined by RNA-sequence were identified to construct a network. From which, 8 significant pathways involved in keloid disorder pathogenesis were enriched and 6 of them were verified. Furthermore, topologically associated domains at susceptible loci were located via the Hi-C database and ten differentially expressed RNAs were identified. Among them, the functions of six molecules for cell proliferation, cell cycle and apoptosis were particularly examined and confirmed by overexpressing and knocking-down assays. This study firstly revealed unknown key biomarkers and pathways in keloid lesions using RNA-sequence and previously reported mutation loci, indicating a feasible approach to reveal the genetic contribution to keloid disorder and possibly to other diseases that are failed by GWAS analysis alone.     

HiCORE: Hi-C Analysis for Identification of Core Chromatin Looping Regions with Higher Resolution

Genome-wide chromosome conformation capture (3C)-based high-throughput sequencing (Hi-C) has enabled identification of genome-wide chromatin loops. Because the Hi-C map with restriction fragment resolution is intrinsically associated with sparsity and stochastic noise, Hi-C data are usually binned at particular intervals; however, the binning method has limited reliability, especially at high resolution. Here, we describe a new method called HiCORE, which provides simple pipelines and algorithms to overcome the limitations of single-layered binning and predict core chromatin regions with three-dimensional physical interactions. In this approach, multiple layers of binning with slightly shifted genome coverage are generated, and interacting bins at each layer are integrated to infer narrower regions of chromatin interactions. HiCORE predicts chromatin looping regions with higher resolution, both in human and Arabidopsis genomes, and contributes to the identification of the precise positions of potential genomic elements in an unbiased manner.     

HPRep: Quantifying Reproducibility in HiChIP and PLAC-Seq Datasets

HiChIP and PLAC-Seq are emerging technologies for studying genome-wide long-range chromatin interactions mediated by the protein of interest, enabling more sensitive and cost-efficient interrogation of protein-centric chromatin conformation. However, due to the unbalanced read distribution introduced by protein immunoprecipitation, existing reproducibility measures developed for Hi-C data are not appropriate for the analysis of HiChIP and PLAC-Seq data. Here, we present HPRep, a stratified and weighted correlation metric derived from normalized contact counts, to quantify reproducibility in HiChIP and PLAC-Seq data. We applied HPRep to multiple real datasets and demonstrate that HPRep outperforms existing reproducibility measures developed for Hi-C data. Specifically, we applied HPRep to H3K4me3 PLAC-Seq data from mouse embryonic stem cells and mouse brain tissues as well as H3K27ac HiChIP data from human lymphoblastoid cell line GM12878 and leukemia cell line K562, showing that HPRep can more clearly separate among pseudo-replicates, real replicates, and non-replicates. Furthermore, in an H3K4me3 PLAC-Seq dataset consisting of 11 samples from four human brain cell types, HPRep demonstrated the expected clustering of data that could not be achieved by existing methods developed for Hi-C data, highlighting the need for a reproducibility metric tailored to HiChIP and PLAC-Seq data.     

Accurate identification of centromere locations in yeast genomes using Hi-C

Centromeres are essential for proper chromosome segregation. Despite extensive research, centromere locations in yeast genomes remain difficult to infer, and in most species they are still unknown. Recently, the chromatin conformation capture assay, Hi-C, has been re-purposed for diverse applications, including de novo genome assembly, deconvolution of metagenomic samples and inference of centromere locations. We describe a method, Centurion, that jointly infers the locations of all centromeres in a single genome from Hi-C data by exploiting the centromeres’ tendency to cluster in three-dimensional space. We first demonstrate the accuracy of Centurion in identifying known centromere locations from high coverage Hi-C data of budding yeast and a human malaria parasite. We then use Centurion to infer centromere locations in 14 yeast species. Across all microbes that we consider, Centurion predicts 89% of centromeres within 5 kb of their known locations. We also demonstrate the robustness of the approach in datasets with low sequencing depth. Finally, we predict centromere coordinates for six yeast species that currently lack centromere annotations. These results show that Centurion can be used for centromere identification for diverse species of yeast and possibly other microorganisms.     

SDImpute: A statistical block imputation method based on cell-level and gene-level information for dropouts in single-cell RNA-seq data

The single-cell RNA sequencing (scRNA-seq) technologies obtain gene expression at single-cell resolution and provide a tool for exploring cell heterogeneity and cell types. As the low amount of extracted mRNA copies per cell, scRNA-seq data exhibit a large number of dropouts, which hinders the downstream analysis of the scRNA-seq data. We propose a statistical method, SDImpute (Single-cell RNA-seq Dropout Imputation), to implement block imputation for dropout events in scRNA-seq data. SDImpute automatically identifies the dropout events based on the gene expression levels and the variations of gene expression across similar cells and similar genes, and it implements block imputation for dropouts by utilizing gene expression unaffected by dropouts from similar cells. In the experiments, the results of the simulated datasets and real datasets suggest that SDImpute is an effective tool to recover the data and preserve the heterogeneity of gene expression across cells. Compared with the state-of-the-art imputation methods, SDImpute improves the accuracy of the downstream analysis including clustering, visualization, and differential expression analysis.     

HiCdat: a fast and easy-to-use Hi-C data analysis tool

Background

The study of nuclear architecture using Chromosome Conformation Capture (3C) technologies is a novel frontier in biology. With further reduction in sequencing costs, the potential of Hi-C in describing nuclear architecture as a phenotype is only about to unfold. To use Hi-C for phenotypic comparisons among different cell types, conditions, or genetic backgrounds, Hi-C data processing needs to be more accessible to biologists.

Results

HiCdat provides a simple graphical user interface for data pre-processing and a collection of higher-level data analysis tools implemented in R. Data pre-processing also supports a wide range of additional data types required for in-depth analysis of the Hi-C data (e.g. RNA-Seq, ChIP-Seq, and BS-Seq).

Conclusions

HiCdat is easy-to-use and provides solutions starting from aligned reads up to in-depth analyses. Importantly, HiCdat is focussed on the analysis of larger structural features of chromosomes, their correlation to genomic and epigenomic features, and on comparative studies. It uses simple input and output formats and can therefore easily be integrated into existing workflows or combined with alternative tools.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-015-0678-x) contains supplementary material, which is available to authorized users.     

The sequencing bias relaxed characteristics of Hi-C derived data and implications for chromatin 3D modeling

The 3D chromatin structure modeling by chromatin interactions derived from Hi-C experiments is significantly challenged by the intrinsic sequencing biases in these experiments. Conventional modeling methods only focus on the bias among different chromatin regions within the same experiment but neglect the bias arising from different experimental sequencing depth. We now show that the regional interaction bias is tightly coupled with the sequencing depth, and we further identify a chromatin structure parameter as the inherent characteristics of Hi-C derived data for chromatin regions. Then we present an approach for chromatin structure prediction capable of relaxing both kinds of sequencing biases by using this identified parameter. This method is validated by intra and inter cell-line comparisons among various chromatin regions for four human cell-lines (K562, GM12878, IMR90 and H1hESC), which shows that the openness of chromatin region is well correlated with chromatin function. This method has been executed by an automatic pipeline (AutoChrom3D) and thus can be conveniently used.     

Spatial organization of the mouse genome and its role in recurrent chromosomal translocations

The extent to which the three-dimensional organization of the genome contributes to chromosomal translocations is an important question in cancer genomics. We generated a high-resolution Hi-C spatial organization map of the G1-arrested mouse pro-B cell genome and used high-throughput genome-wide translocation sequencing to map translocations from target DNA double-strand breaks (DSBs) within it. RAG endonuclease-cleaved antigen-receptor loci are dominant translocation partners for target DSBs regardless of genomic position, reflecting high-frequency DSBs at these loci and their colocalization in a fraction of cells. To directly assess spatial proximity contributions, we normalized genomic DSBs via ionizing radiation. Under these conditions, translocations were highly enriched in cis along single chromosomes containing target DSBs and within other chromosomes and subchromosomal domains in a manner directly related to pre-existing spatial proximity. By combining two high-throughput genomic methods in a genetically tractable system, we provide a new lens for viewing cancer genomes.     

Cluster analysis and disease mapping--why,when, and how? A step by step guide.

Growing public awareness of environmental hazards has led to an increased demand for public health authorities to investigate geographical clustering of diseases. Although such cluster analysis is nearly always ineffective in identifying causes of disease, it often has to be used to address public concern about environmental hazards. Interpreting the resulting data is not straightforward, however, and this paper presents a guide for the non-specialist. The pitfalls include the fact that cluster analyses are usually done post hoc, and not as a result of a prior hypothesis. This is particularly true for investigations prompted by reported clusters, which have the inherent danger of overestimating the disease rate through "boundary shrinkage" of the population from which the cases are assumed to have arisen. In disease surveillance the problem of making multiple comparisons can be overcome by testing for clustering and autocorrelation. When rates of disease are illustrated in disease maps undue focus on areas where random fluctuation is greatest can be minimised by smoothing techniques. Despite the fact that cluster analyses rarely prove fruitful in identifying causation, they may-like single case reports-have the potential to generate new knowledge.     

A comparison of cluster analysis methods using DNA methylation data

MOTIVATION: Aberrant DNA methylation is common in cancer. DNA methylation profiles differ between tumor types and subtypes and provide a powerful diagnostic tool for identifying clusters of samples and/or genes. DNA methylation data obtained with the quantitative, highly sensitive MethyLight technology is not normally distributed; it frequently contains an excess of zeros. Established tools to analyze this type of data do not exist. Here, we evaluate a variety of methods for cluster analysis to determine which is most reliable. RESULTS: We introduce a Bernoulli-lognormal mixture model for clustering DNA methylation data obtained using MethyLight. We model the outcomes using a two-part distribution having discrete and continuous components. It is compared with standard cluster analysis approaches for continuous data and for discrete data. In a simulation study, we find that the two-part model has the lowest classification error rate for mixture outcome data compared with other approaches. The methods are illustrated using DNA methylation data from a study of lung cancer cell lines. Compared with competing hierarchical clustering methods, the mixture model approaches have the lowest cross-validation error for detecting lung cancer subtype (non-small versus small cell). The Bernoulli-lognormal mixture assigns observations to subgroups with the lowest uncertainty. AVAILABILITY: Software is available upon request from the authors. SUPPLEMENTARY INFORMATION: http://www-rcf.usc.edu/~kims/SupplementaryInfo.html