FIREcaller: Detecting Frequently Interacting Regions from Hi-C Data

Abstract

MotivationHi-C experiments have been widely adopted to study chromatin spatial organization, which plays an important role in genome function. Well-established Hi-C readouts include A/B compartments, topologically associating domains (TADs) and chromatin loops. We have recently proposed another readout: frequently interacting regions (FIREs) and discovered them to be informative about tissue-specific gene expression. However, computational tools for detecting FIREs from Hi-C data are still lacking.ResultsIn this work, we have developed FIREcaller, a stand-alone, user-friendly R package for detecting FIREs from Hi-C data. FIREcaller takes raw Hi-C contact matrix as input, performs within-sample and cross-sample normalization via HiCNormCis and quantile normalization respectively, and outputs continuous FIRE scores, dichotomous FIREs and super-FIREs.Availability and implementationThe FIREcaller package is implemented in R, freely available athttps://yunliweb.its.unc.edu/FIREcaller.Contactyunli@med.unc.eduorhum@ccf.org

Neuronal and glial 3D chromatin architecture illustrates cellular etiology of brain disorders

“…We therefore compared FIREs in NeuN+ and NeuN− cells to identify how local chromatin architecture differs among major brain cell types. We detected 3,966 and 3,967 FIREs in NeuN+ and NeuN− cells, respectively, with slightly fewer than 40% of the FIREs (n=1,499) shared between both samples (Figure 1a) 26 . To further investigate how FIREs are associated with cell-type-specific gene expression profiles, we used a stringent cutoff to define differential FIREs on the basis of the FIRE score (Methods), detecting 287 differential FIREs between NeuN+ (145) and NeuN− (142) cells (hereby referred to as NeuN+ and NeuN− FIREs, respectively, Figure 1b, Supplementary Table 1).…”

“…FIREs represent regions that act as interaction hubs 6,26 . They are enriched with regulatory elements, suggesting that chromatin interactome may provide regulatory regions.…”

“…We used FIREcaller (v.1.10) 26 to define FIREs and super FIREs. NeuN+ differential FIREs were identified by obtaining the genomic regions that have NeuN+ FIRE scores greater than qnorm (0.975) and NeuN− FIRE scores lower than qnorm (0.9).…”

Nucleotide excision repair hotspots and coldspots of UV-induced DNA damage in the human genome

“…To test this hypothesis, we sought to interrogate the publicly available Hi-C data of human fibroblast cell line IMR90 (35,36). Specifically, after quality control procedures and data normalization, we profiled frequently interacting regions (FIREs) using FIREcaller (37). After overlapping the repair hotspots and coldspots with the called FIREs (Table S7A), we found that a significantly higher proportion of repair hotspots overlap with FIREs -23.16% and 11.76% for (6-4)PP and CPD, respectivelycompared to a genome average of 6.93% based on the profiled FIREs ( Figure 6A).…”

“…We adopted the Hi-C data of human fibroblast cell line IMR90 (35,36) to investigate the relationship between identified repair hotspots and the 3D genome structure. We took the raw contact matrix with 40 kb resolution as input and detected FIREs, which play important roles in transcriptional regulations, across the entire genome using FIREcaller (37). To further investigate whether these repair hotspots are involved in functional chromatin looping between regulatory elements and their target genes, we adopted the Fit-Hi-C approach (40) to identify long-range chromatin interactions on all 40 kb bin pairs within a maximal 3 MB region.…”