Modeling and analysis of Hi-C data by HiSIF identifies characteristic promoter-distal loops

Abstract

Current computational methods on Hi-C analysis focused on identifying Mb-size domains often failed to unveil the underlying functional and mechanistic relationship of chromatin structure and gene regulation. We developed a novel computational method HiSIF to identify genome-wide interacting loci. We illustrated HiSIF outperformed other tools for identifying chromatin loops. We applied it to Hi-C data in breast cancer cells and identified 21 genes with gained loops showing worse relapse-free survival in endocrine-treated patients, suggesting the genes with enhanced loops can be used for prognostic signatures for measuring the outcome of the endocrine treatment. HiSIF is available at https://github.com/yufanzhouonline/HiSIF.

Integrating whole genome sequencing, methylation, gene expression, topologically associated domain information in regulatory mutation prediction: A study of follicular lymphoma

“…Annotation to the reference genome were performed on four defined genomic regions: the TSS/TES regions between −5kb and + 1kp to the TSS (transcription start site)/TES (transcription end site) of protein-coding genes, and the gene body region between TSS and TES, and the 5′distance regions were calculated from 1 Mb to 5 kb upstream of the TSS. These four defined regions are similar to the previous publications [50] , [51] in differential methylation analysis and identification of promoter-distal loops. Gene expression levels are measured as reads per kilobase of transcript per million mapped reads (RPKM) of RNA-Seq experiments and were computed by applying the featureCounts [52] and our in-house Python code on aligned BAM files.…”

Hi-C profiling of cancer spheroids identifies 3D-growth-specific chromatin interactions in breast cancer endocrine resistance

“…We applied HiSIF [ 39 ] to identify significant interaction fragments (SIFs) for the Hi-C data in both 2D monolayers and 3D spheroids. At the optimal parameters set (t = 1, 20 Kb, FDR = 0.1), we obtained a total of 577,585 for MCF7_2D, 431,056 for MCF7_3D, 327,064 for MCF7TR_2D and 319,644 for MCF7TR_3D, respectively.…”

“…We used HiSIF [ 39 ] to identify significant interaction fragments (SIFs) with all valid pairs from HiC-Pro [ 35 ] and HISIF parameters of t = 1, s = 2, p = 1 29, w = 50, 500, 20,000. We further used FDR ≤ 0.1 as the cutoff to select the final set of SIFs.…”

Mapping nucleosome and chromatin architectures: A survey of computational methods

“…For example, NPS/iNPS [24] , [25] applied a gaussian convolution algorithm to detect inflection points to find candidate nucleosomes, and a novel iterative correction (ICE) algorithm [26] was developed for balancing the biases in Hi-C data. The computational algorithms are often further implemented as software tools for user-friendly and interactive interfaces or downloadable executable files in public code repositories, such as NucHMM [27] for identifying nucleosome states, JuiceBox [28] for visualizing and analyzing Hi-C data, and HiSIF [29] for detecting significant interacting fragments. In addition, a series of methods and tools are sometimes combined into a workflow/pipeline to achieve its meaningful biological output.…”