In terms of in NPC, the structure information of 129S1 allele was represented by V6.5 icSHAPE data, since they have the same sequence. Other published datasets used in this study are listed as follows: (1) “type”:”entrez-geo”,”attrs”:”text”:”GSE69143″,”term_id”:”69143″GSE69143: mouse ChIRP-seq profile [45]; (2) “type”:”entrez-geo”,”attrs”:”text”:”GSE102518″,”term_id”:”102518″GSE102518: mouse V6.5 ESC ChIP-seq CFTRinh-172 data of H3K4me1, H3K4me3, H3K27ac, H3K27me3, and H3K9me3 [37]; (3) “type”:”entrez-geo”,”attrs”:”text”:”GSE117289″,”term_id”:”117289″GSE117289: mouse NPC ChIP-seq data of H3K4me1, H3K4me3, H3K27ac, and H3K27me3 [79]; (4) mouse V6.5 ESC icSHAPE data from the whole cell [61]; “type”:”entrez-geo”,”attrs”:”text”:”GSE64169″,”term_id”:”64169″GSE64169 and cell compartments [65] (“type”:”entrez-geo”,”attrs”:”text”:”GSE117840″,”term_id”:”117840″GSE117840); (5) “type”:”entrez-geo”,”attrs”:”text”:”GSE52681″,”term_id”:”52681″GSE52681: mouse ESC m6A sequencing data [68]; (5) “type”:”entrez-geo”,”attrs”:”text”:”GSE82312″,”term_id”:”82312″GSE82312: GRID-seq profiles from human ES cell lines MM1S & MDA231 and mouse ESC [20]; (6) “type”:”entrez-geo”,”attrs”:”text”:”GSE92345″,”term_id”:”92345″GSE92345: MARGI profiles from human ES cell lines H9 [21]; (7) “type”:”entrez-geo”,”attrs”:”text”:”GSE66478″,”term_id”:”66478″GSE66478: biochemical fractionation of HEK293 nuclei and RNA-seq of chromatin-associated and soluble-nuclear RNA [19]; (8) “type”:”entrez-geo”,”attrs”:”text”:”GSE21227″,”term_id”:”21227″GSE21227: chromatin-associated RNAs (CARs) from human fibroblast (HF) cells [17]; (9) “type”:”entrez-geo”,”attrs”:”text”:”GSE57231″,”term_id”:”57231″GSE57231: total RNA-seq profiles of mouse V6.5 ESC [80]; (10) “type”:”entrez-geo”,”attrs”:”text”:”GSE32916″,”term_id”:”32916″GSE32916: subcellular RNA-seq profiles of mouse V6.5 ESC [18]; (11) All RNA binding peaks in ChIRP/CHART/RAP/GRID-seq experiments were downloaded from LnChrom [43]. Abstract We develop PIRCh-seq, a method which enables a comprehensive survey of chromatin-associated RNAs in a Mmp11 histone modification-specific manner. We identify hundreds of chromatin-associated RNAs in several cell types with substantially less contamination by nascent transcripts. Non-coding RNAs are found enriched on chromatin and are classified into functional groups based on the patterns of their association with specific histone modifications. We find single-stranded RNA bases are more chromatin-associated, and we discover hundreds of allele-specific RNA-chromatin interactions. These results provide a unique resource to globally study the functions of chromatin-associated lncRNAs and elucidate the basic mechanisms of chromatin-RNA interactions. Introduction RNAs are both the product of transcription and major regulators of the transcriptional process. In particular, long non-coding RNAs (lncRNAs) are numerous in eukaryotes and function in many cases as transcription regulators [1C3]. With the development of next-generation sequencing (NGS), tens of thousands of lncRNAs have been revealed in both murine and human genomes, and have emerged as important regulators for different biological processes [4, 5]. However, among all expressed lncRNAs, only a small subset are shown to be cell essential [6] or important for development [7] or immune responses [8]. Strategies to annotate biochemical properties of lncRNAs will be helpful to prioritize lncRNA candidates for functional analyses. Some well-studied cases have indicated that one major mechanism of lncRNAs is usually their ability to function through binding to histone-modifying complexes [9, 10]. LncRNAs can either recruit chromatin modifiers to regulate the chromatin says or directly regulate the process of transcription through chromosome looping to bridge distal enhancer elements to promoters [11, 12]. Thereby, a genome-wide identification of chromatin-associated lncRNAs may reveal functions and mechanisms of lncRNAs in mediating chromatin modification CFTRinh-172 and regulating gene transcription. A considerable amount of literature has been published concerning protein-RNA interactions. The advent of technologies such as RIP [13], CLIP [14], fRIP [15], and CARIP [16] has led to the discovery of multiple protein-associated RNAs, including many chromatin regulators. Conversely, nuclear extraction methods followed by RNA-seq have enabled the detection of lncRNAs which are physically associated with chromatin [17C19]. In addition, more recently reported methods like GRID-seq [20], MARGI [21], and SPRITE [22] can be used to capture pairwise RNA interactions with DNA. However, these approaches are not capable of revealing which chromatin modifications are associated with specific lncRNAs and are thus limited in the ability to elucidate their potential regulatory functions. For instance, a large CFTRinh-172 number of lncRNAs are associated with Polycomb Repressive Complex 2 (PRC2), a key mammalian epigenetic regulator, to silence gene transcription by targeting its genomic loci and trimethylating histone H3 lysine 27 (H3K27me3) [23]. Therefore, lncRNAs associated with PRC2 complex may be enriched on heterochromatin regions with H3K27me3 modification. On the other hand, a new class of lncRNAs called super-lncRNAs was recently characterized. These lncRNAs target super-enhancers which have potential to regulate enhancer activities and transcription [24]. These super-lncRNAs may be enriched on euchromatin and active DNA regulatory elements with histone H3 lysine 27 acetylation (H3K27ac), H3 lysine 4 monomethylation (H3K4me1), and trimethylation (H3K4me3). Therefore, we believe it will.
Categories