RNA Polymerase II Research Articles

Temporal-iCLIP captures co-transcriptional RNA-protein interactions

Dynamic RNA-protein interactions govern the co-transcriptional packaging of RNA polymerase II (RNAPII)-derived transcripts. Yet, our current understanding of this process in vivo primarily stems from steady state analysis. To remedy this, we here conduct temporal-iCLIP (tiCLIP), combining RNAPII transcriptional synchronisation with UV cross-linking of RNA-protein complexes at serial timepoints. We apply tiCLIP to the RNA export adaptor, ALYREF; a component of the Nuclear Exosome Targeting (NEXT) complex, RBM7; and the nuclear cap binding complex (CBC). Regardless of function, all tested factors interact with nascent RNA as it exits RNAPII. Moreover, we demonstrate that the two transesterification steps of pre-mRNA splicing temporally separate ALYREF and RBM7 binding to splicing intermediates, and that exon-exon junction density drives RNA 5′end binding of ALYREF. Finally, we identify underappreciated steps in snoRNA 3′end processing performed by RBM7. Altogether, our data provide a temporal view of RNA-protein interactions during the early phases of transcription.

Elongation factor 1 is a component of the ArabidopsisRNA polymerase II elongation complex and associates with a subset of transcribed genes.

Elongation factors modulate the efficiency of mRNA synthesis by RNA polymerase II (RNAPII) in the context of chromatin, thus contributing to implement proper gene expression programmes. The zinc-finger protein elongation factor 1 (ELF1) is a conserved transcript elongation factor (TEF), whose molecular function so far has not been studied in plants. Using biochemical approaches, we examined the interaction of Arabidopsis ELF1 with DNA and histones invitro and with the RNAPII elongation complex invivo. In addition, cytological assays demonstrated the nuclear localisation of the protein, and by means of double-mutant analyses, interplay with genes encoding other elongation factors was explored. The genome-wide distribution of ELF1 was addressed by chromatin immunoprecipitation. ELF1 isolated from Arabidopsis cells robustly copurified with RNAPII and various other elongation factors including SPT4-SPT5, SPT6, IWS1, FACT and PAF1C. Analysis of a CRISPR-Cas9-mediated gene editing mutant of ELF1 revealed distinct genetic interactions with mutants deficient in other elongation factors. Moreover, ELF1 associated with genomic regions actively transcribed by RNAPII. However, ELF1 occupied only c. 33% of the RNAPII transcribed loci with preference for inducible rather than constitutively expressed genes. Collectively, these results establish that Arabidopsis ELF1 shares several characteristic attributes with RNAPII TEFs.

Transcription-induced active forces suppress chromatin motion by inducing a transient disorder-to-order transition.

Recent experiments have shown that the mobility of human interphase chromosome decreases during transcription, and increases upon inhibiting transcription, a finding that is counter-intuitive because it is thought that the active mechanical force ($F$) generated by RNA polymerase II (RNAPII) on chromatin would render it more open and mobile. Inspired by these observations, we use a copolymer model to investigate how $F$ affects the dynamical properties of a single chromatin. The movements of the loci in the gene-rich region are suppressed in an intermediate range of $F$, and are enhanced at small and large $F$ values. In the intermediate $F$, the bond length between consecutive loci increases, becoming commensurate with the location of the minimum in the attractive interaction between the active loci. This results in a transient disorder-to-order transition, leading to the decreased mobility during transcription. Transient ordering of the loci in the gene-rich region might be a mechanism for nucleating a dynamic network involving transcription factors, RNAPII, and chromatin.

Uncovering the conditions that dictate the liquid-liquid phase separation of the CTD of RNAPII.

The intrinsically disordered carboxy terminal domain (CTD) of RNA polymerase II (RNAPII) may nucleate the formation of transcription factories through a mechanism of liquid-liquid phase separation (LLPS). Transcription factories are densely packed membraneless organelles (MLOs) that contain all the necessary components to perform the transcription cycle. The CTD in model organisms such as S. cerevisiae, S. pombe and C. albicans, is composed of ∼26 copies of consensus repeats (Y1S2P3T4S5P6S7). However, CTD sequences of fungi isolated from geographically distinct locations from temperate to polar regions, show high sequence diversity at positions 1, 4, and 7. The CTD is highly conserved at its Ser-Pro motifs, which are targets of phosphorylation and cis-trans prolyl isomerization by Ess1. Ess1, an essential peptidylprolyl isomerase in yeast, has a highly conserved evolutionary relationship with the CTD. Here, we found that different CTDs isolated from yeasts found in geographically distinct regions ranging from temperate to polar climates undergo LLPS, but do so differently. We hypothesize Ess1 and phosphorylation may modulate the phase behavior of the CTD. The objective of this study is to determine how linear amino acid sequence, phosphorylation, and cis-trans prolyl isomerization influences CTD phase behavior. To do so, we utilize a variety of biophysical characterization assays such as UV-Vis spectroscopy, DIC microscopy, and phase separation prediction algorithms. These results will elucidate what conditions promote the LLPS of the CTD and its orthologs, thereby setting the stage for further investigating the conditions that promote the formation of transcription factories.

A New Insight into MYC Action: Control of RNA Polymerase II Methylation and Transcription Termination

MYC oncoprotein deregulation is a common catastrophic event in human cancer and limiting its activity restrains tumor development and maintenance, as clearly shown via Omomyc, an MYC-interfering 90 amino acid mini-protein. MYC is a multifunctional transcription factor that regulates many aspects of transcription by RNA polymerase II (RNAPII), such as transcription activation, pause release, and elongation. MYC directly associates with Protein Arginine Methyltransferase 5 (PRMT5), a protein that methylates a variety of targets, including RNAPII at the arginine residue R1810 (R1810me2s), crucial for proper transcription termination and splicing of transcripts. Therefore, we asked whether MYC controls termination as well, by affecting R1810me2S. We show that MYC overexpression strongly increases R1810me2s, while Omomyc, an MYC shRNA, or a PRMT5 inhibitor and siRNA counteract this phenomenon. Omomyc also impairs Serine 2 phosphorylation in the RNAPII carboxyterminal domain, a modification that sustains transcription elongation. ChIP-seq experiments show that Omomyc replaces MYC and reshapes RNAPII distribution, increasing occupancy at promoter and termination sites. It is unclear how this may affect gene expression. Transcriptomic analysis shows that transcripts pivotal to key signaling pathways are both up- or down-regulated by Omomyc, whereas genes directly controlled by MYC and belonging to a specific signature are strongly down-regulated. Overall, our data point to an MYC/PRMT5/RNAPII axis that controls termination via RNAPII symmetrical dimethylation and contributes to rewiring the expression of genes altered by MYC overexpression in cancer cells. It remains to be clarified which role this may have in tumor development.

Mapping Multi-factor-mediated Chromatin Interactions to Assess Dysregulation of Lung Cancer-related Genes

Studies on the lung cancer genome are indispensable for developing a cure for lung cancer. Whole-genome resequencing, genome-wide association studies, and transcriptome sequencing have greatly improved our understanding of the cancer genome. However, dysregulation of long-range chromatin interactions in lung cancer remains poorly described. To better understand the three-dimensional (3D) genomic interaction features of the lung cancer genome, we used the A549 cell line as a model system and generated high-resolution chromatin interactions associated with RNA polymerase II (RNAPII), CCCTC-binding factor (CTCF), enhancer of zeste homolog 2 (EZH2), and histone 3 lysine 27 trimethylation (H3K27me3) using long-read chromatin interaction analysis by paired-end tag sequencing (ChIA-PET). Analysis showed that EZH2/H3K27me3-mediated interactions further repressed target genes, either through loops or domains, and their distributions along the genome were distinct from and complementary to those associated with RNAPII. Cancer-related genes were highly enriched with chromatin interactions, and chromatin interactions specific to the A549 cell line were associated with oncogenes and tumor suppressor genes, such as additional repressive interactions on FOXO4 and promoter–promoter interactions between NF1 and RNF135. Knockout of an anchor associated with chromatin interactions reversed the dysregulation of cancer-related genes, suggesting that chromatin interactions are essential for proper expression of lung cancer-related genes. These findings demonstrate the 3D landscape and gene regulatory relationships of the lung cancer genome.

Active mRNA degradation by EXD2 nuclease elicits recovery of transcription after genotoxic stress

The transcriptional response to genotoxic stress involves gene expression arrest, followed by recovery of mRNA synthesis (RRS) after DNA repair. We find that the lack of the EXD2 nuclease impairs RRS and decreases cell survival after UV irradiation, without affecting DNA repair. Overexpression of wild-type, but not nuclease-dead EXD2, restores RRS and cell survival. We observe that UV irradiation triggers the relocation of EXD2 from mitochondria to the nucleus. There, EXD2 is recruited to chromatin where it transiently interacts with RNA Polymerase II (RNAPII) to promote the degradation of nascent mRNAs synthesized at the time of genotoxic attack. Reconstitution of the EXD2-RNAPII partnership on a transcribed DNA template in vitro shows that EXD2 primarily interacts with an elongation-blocked RNAPII and efficiently digests mRNA. Overall, our data highlight a crucial step in the transcriptional response to genotoxic attack in which EXD2 interacts with elongation-stalled RNAPII on chromatin to potentially degrade the associated nascent mRNA, allowing transcription restart after DNA repair.

Epigenetic Regulation of HIV-1 Sense and Antisense Transcription in Response to Latency-Reversing Agents.

Nucleosomes positioned on the HIV-1 5' long terminal repeat (LTR) regulate sense transcription as well as the establishment and maintenance of latency. A negative-sense promoter (NSP) in the 3' LTR expresses antisense transcripts with coding and non-coding activities. Previous studies identified cis-acting elements that modulate NSP activity. Here, we used the two chronically infected T cell lines, ACH-2 and J1.1, to investigate epigenetic regulation of NSP activity. We found that histones H3 and H4 are present on the 3' LTR in both cell lines. Following treatment with histone deacetylase inhibitors (HDACi), the levels of H3K27Ac increased and histone occupancy declined. HDACi treatment also led to increased levels of RNA polymerase II (RNPII) at NSP, and antisense transcription was induced with similar kinetics and to a similar extent as 5' LTR-driven sense transcription. We also detected H3K9me2 and H3K27me3 on NSP, along with the enzymes responsible for these epigenetic marks, namely G9a and EZH2, respectively. Treatment with their respective inhibitors had little or no effect on RNPII occupancy at the two LTRs, but it induced both sense and antisense transcription. Moreover, the increased expression of antisense transcripts in response to treatment with a panel of eleven latency-reversing agents closely paralleled and was often greater than the effect on sense transcripts. Thus, HIV-1 sense and antisense RNA expression are both regulated via acetylation and methylation of lysine 9 and 27 on histone H3. Since HIV-1 antisense transcripts act as non-coding RNAs promoting epigenetic silencing of the 5' LTR, our results suggest that the limited efficacy of latency-reversing agents in the context of 'shock and kill' cure strategies may be due to concurrent induction of antisense transcripts thwarting their effect on sense transcription.

Transcription coupled DNA repair protein UVSSA binds to DNA and RNA: Mapping of nucleic acid interaction sites on human UVSSA

Transcription-coupled repair (TCR) is a dedicated pathway for the preferential repair of bulky transcription-blocking DNA lesions. These lesions stall the elongating RNA-polymerase II (RNAPII) triggering the recruitment of TCR proteins at the damaged site. UV-stimulated scaffold protein A (UVSSA) is a recently identified cofactor which is involved in stabilization of the TCR complex, recruitment of DNA-repair machinery and removal/restoration of RNAPII from the lesion site. Mutations in UVSSA render the cells TCR-deficient and have been linked to UV-sensitive syndrome. Human UVSSA is a 709-residue long protein with two short conserved domains; an N-terminal (residues 1–150) and a C-terminal (residues 495–605) domain, while the rest of the protein is predicted to be intrinsically disordered. The protein is well conserved in eukaryotes, however; none of its homologs have been characterized yet. Here, we have purified the recombinant human UVSSA and have characterized it using bioinformatics, biophysical and biochemical techniques. Using EMSA, SPR and fluorescence-based methods, we have shown that human UVSSA interacts with DNA and RNA. Furthermore, we have mapped the nucleic acid binding regions using several recombinant protein fragments containing either the N-terminal or the C-terminal domains. Our data indicate that UVSSA possesses at least two nucleic acid binding regions; the N-terminal domain and a C-terminal tail region (residues 606–662). These regions, far apart in sequence space, are predicted to be in close proximity in structure-space suggesting a coherent interaction with target DNA/RNA. The study may provide functional clues about the novel family of UVSSA proteins.

Sumoylation is Largely Dispensable for Normal Growth but Facilitates Heat Tolerance in Yeast.

Numerous proteins are sumoylated in normally growing yeast and SUMO conjugation levels rise upon exposure to several stress conditions. We observe high levels of sumoylation also during early exponential growth and when nutrient-rich medium is used. However, we find that reduced sumoylation (∼75% less than normal) is remarkably well-tolerated, with no apparent growth defects under nonstress conditions or under osmotic, oxidative, or ethanol stresses. In contrast, strains with reduced activity of Ubc9, the sole SUMO conjugase, are temperature-sensitive, implicating sumoylation in the heat stress response, specifically. Aligned with this, a mild heat shock triggers increased sumoylation which requires functional levels of Ubc9, but likely also depends on decreased desumoylation, since heat shock reduces protein levels of Ulp1, the major SUMO protease. Furthermore, we find that a ubc9 mutant strain with only ∼5% of normal sumoylation levels shows a modest growth defect, has abnormal genomic distribution of RNA polymerase II (RNAPII), and displays a greatly expanded redistribution of RNAPII after heat shock. Together, our data implies that SUMO conjugations are largely dispensable under normal conditions, but a threshold level of Ubc9 activity is needed to maintain transcriptional control and to modulate the redistribution of RNAPII and promote survival when temperatures rise.

Analysis of RNA Polymerase II Chromatin Binding by Flow Cytometry.

RNA polymerase II (RNAPII) transcribes DNA into mRNA and thereby plays a critical role in cellular protein production. In addition, RNAPII plays a central role in DNA damage responses. Measurements of RNAPII on chromatin may thus give insight into several essential processes in eukaryotic cells. During transcription, the C-terminal domain of RNAPII becomes post-translationally modified, and phosphorylation on serine 5 and serine 2 can be used as markers for the promoter proximal and productively elongating forms of RNAPII, respectively. Here, we provide a detailed protocol for the detection of chromatin-bound RNAPII and its serine 5- and serine 2-phosphorylated forms in individual human cells through the cell cycle. We have recently shown that this method can be used to study the effects of ultraviolet DNA damage on RNAPII chromatin binding and that it can even be used to reveal new knowledge about the transcription cycle itself. Other commonly used methods to study RNAPII chromatin binding include chromatin immunoprecipitation followed by sequencing or chromatin fractionation followed by western blotting. However, such methods are frequently based on lysates made from a large number of cells, which may mask population heterogeneity, e.g., due to cell cycle phase. With strengths such as single-cell analysis, speed of use, and accurate quantitative readouts, we envision that our flow cytometry method can be widely used as a complementary approach to sequencing-based methods to study effects of different stimuli and inhibitors on RNAPII-mediated transcription. Graphical overview.

MDC1 maintains active elongation complexes of RNA polymerase II

The role of MDC1 in the DNA damage response has been extensively studied; however, its impact on other cellular processes is not well understood. Here, we describe the role of MDC1 in transcription as a regulator of RNA polymerase II (RNAPII). Depletion of MDC1 causes a genome-wide reduction in the abundance of actively engaged RNAPII elongation complexes throughout the gene body of protein-encoding genes under unperturbed conditions. Decreased engaged RNAPII subsequently alters the assembly of the spliceosome complex on chromatin, leading to changes in pre-mRNA splicing. Mechanistically, the S/TQ domain of MDC1 modulates RNAPII-mediated transcription. Upon genotoxic stress, MDC1 promotes the abundance of engaged RNAPII complexes at DNA breaks, thereby stimulating nascent transcription at the damaged sites. Of clinical relevance, cancer cells lacking MDC1 display hypersensitivity to RNAPII inhibitors. Overall, we unveil a role of MDC1 in RNAPII-mediated transcription with potential implications for cancer treatment.

The Integrator complex desensitizes cellular response to TGF-β/BMP signaling

Maintenance of stem cells requires the concerted actions of niche-derived signals and stem cell-intrinsic factors. Although Decapentaplegic (Dpp), a Drosophila bone morphogenetic protein (BMP) molecule, can act as a long-range morphogen, its function is spatially limited to the germline stem cell niche in the germarium. We show here that Integrator, a complex known to be involved in RNA polymerase II (RNAPII)-mediated transcriptional regulation in the nucleus, promotes germline differentiation by restricting niche-derived Dpp/BMP activity in the cytoplasm. Further results show that Integrator works in various developmental contexts to desensitize the cellular response to Dpp/BMP signaling during Drosophila development. Mechanistically, our results show that Integrator forms a multi-subunit complex with the type I receptor Thickveins (Tkv) and other Dpp/BMP signaling components and acts in a negative feedback loop to promote Tkv turnover independent of its transcriptional activity. Similarly, human Integrator subunits bind transforming growth factor β (TGF-β)/BMP signaling components and antagonize their activity, suggesting a conserved role of Integrator across metazoans.

Polyglutamine-expanded ATXN7 alters a specific epigenetic signature underlying photoreceptor identity gene expression in SCA7 mouse retinopathy

BackgroundSpinocerebellar ataxia type 7 (SCA7) is a neurodegenerative disorder that primarily affects the cerebellum and retina. SCA7 is caused by a polyglutamine expansion in the ATXN7 protein, a subunit of the transcriptional coactivator SAGA that acetylates histone H3 to deposit narrow H3K9ac mark at DNA regulatory elements of active genes. Defective histone acetylation has been presented as a possible cause for gene deregulation in SCA7 mouse models. However, the topography of acetylation defects at the whole genome level and its relationship to changes in gene expression remain to be determined.MethodsWe performed deep RNA-sequencing and chromatin immunoprecipitation coupled to high-throughput sequencing to examine the genome-wide correlation between gene deregulation and alteration of the active transcription marks, e.g. SAGA-related H3K9ac, CBP-related H3K27ac and RNA polymerase II (RNAPII), in a SCA7 mouse retinopathy model.ResultsOur analyses revealed that active transcription marks are reduced at most gene promoters in SCA7 retina, while a limited number of genes show changes in expression. We found that SCA7 retinopathy is caused by preferential downregulation of hundreds of highly expressed genes that define morphological and physiological identities of mature photoreceptors. We further uncovered that these photoreceptor genes harbor unusually broad H3K9ac profiles spanning the entire gene bodies and have a low RNAPII pausing. This broad H3K9ac signature co-occurs with other features that delineate superenhancers, including broad H3K27ac, binding sites for photoreceptor specific transcription factors and expression of enhancer-related non-coding RNAs (eRNAs). In SCA7 retina, downregulated photoreceptor genes show decreased H3K9 and H3K27 acetylation and eRNA expression as well as increased RNAPII pausing, suggesting that superenhancer-related features are altered.ConclusionsOur study thus provides evidence that distinctive epigenetic configurations underlying high expression of cell-type specific genes are preferentially impaired in SCA7, resulting in a defect in the maintenance of identity features of mature photoreceptors. Our results also suggest that continuous SAGA-driven acetylation plays a role in preserving post-mitotic neuronal identity.

Approach to Measuring the Effect of PARP1 on RNA Polymerase II Elongation Rates.

The rate of RNA polymerase II (RNAPII) transcriptional elongation plays a critical role in mRNA biogenesis, from transcription initiation to alternative splicing. As RNAPII moves along the DNA, it must read the DNA sequences wrapped up as chromatin. Thus, the structure of chromatin impedes the movement and speed at which RNAPII moves, presenting a crucial regulation to gene expression. Therefore, factors that bind and regulate the structure of chromatin will impact the rate of RNAPII elongation. We previously showed that PARP1 (poly-ADP-ribose polymerase 1) is one of such factors that bind and alter chromatin dynamics. We also showed that its alteration of chromatin structure modulates RNAPII processivity during transcriptional elongation. Here, we aim to understand how PARP1 alters RNAPII elongation kinetics genome wide.

Histone H2Bub dynamics in the 5′ region of active genes are tightly linked to the UV-induced transcriptional response

The timing and location of writing and erasing of histone modifications determine gene expression programs and are tightly controlled processes. One such modification is the monoubiquitination of histone H2B (H2Bub), whose precise level during transcription elongation is dynamically regulated by the synergistic action of RNF20/40 ubiquitin-ligase and the de-ubiquitinase (DUB) of the ATXN7L3-containing DUB modules. Here, we characterize the dynamics of H2Bub in transcription and explore its role in perspective with the recently updated model of UV damage-induced transcription reorganization. Employing integrative analysis of genome-wide high-throughput approaches, transcription inhibitors and ATXN7L3-DUB knockdown cells, we find that H2Bub levels and patterns depend on intron-exon architecture both in steady state and upon UV. Importantly, our analysis reveals a widespread redistribution of this histone mark, rather than a uniform loss as previously suggested, which closely mirrors the post-UV dynamics of elongating RNA Polymerase II (RNAPII) at transcribed loci. The observed effects are due to a direct inter-dependence on RNAPII local concentration and speed, and we show that deficient ATXN7L3-mediated DUB activity leads to increased elongation rates in both non-irradiated and irradiated conditions. Our data and the implementation of a high-resolution computational framework reveal that the H2Bub pattern follows that of RNAPII, both in the ATXNL3 knockdown and in response to UV guaranteeing faithful elongation speed, especially in the context of the transcription-driven DNA damage response.

TFIIS Is Crucial During Early Transcript Elongation for Transcriptional Reprogramming in Response to Heat Stress

In addition to the stage of transcriptional initiation, the production of mRNAs is regulated during elongation. Accordingly, the synthesis of mRNAs by RNA polymerase II (RNAPII) in the chromatin context is modulated by various transcript elongation factors. TFIIS is an elongation factor that stimulates the transcript cleavage activity of RNAPII to reactivate stalled elongation complexes at barriers to transcription including nucleosomes. Since Arabidopsis tfIIs mutants grow normally under standard conditions, we have exposed them to heat stress (HS), revealing that tfIIs plants are highly sensitive to elevated temperatures. Transcriptomic analyses demonstrate that particularly HS-induced genes are expressed at lower levels in tfIIs than in wildtype. Mapping the distribution of elongating RNAPII uncovered that in tfIIs plants RNAPII accumulates at the +1 nucleosome of genes that are upregulated upon HS. The promoter-proximal RNAPII accumulation in tfIIs under HS conditions conforms to that observed upon inhibition of the RNAPII transcript cleavage activity. Further analysis of the RNAPII accumulation downstream of transcriptional start sites illustrated that RNAPII stalling occurs at +1 nucleosomes that are depleted in the histone variant H2A.Z upon HS. Therefore, assistance of early transcript elongation by TFIIS is required for reprogramming gene expression to establish plant thermotolerance.

ERK-mediated NELF-A phosphorylation promotes transcription elongation of immediate-early genes by releasing promoter-proximal pausing of RNA polymerase II

Growth factor-induced, ERK-mediated induction of immediate-early genes (IEGs) is crucial for cell growth and tumorigenesis. Although IEG expression is mainly regulated at the level of transcription elongation by RNA polymerase-II (Pol-II) promoter-proximal pausing and its release, the role of ERK in this process remains unknown. Here, we identified negative elongation factor (NELF)-A as an ERK substrate. Upon growth factor stimulation, ERK phosphorylates NELF-A, which dissociates NELF from paused Pol-II at the promoter-proximal regions of IEGs, allowing Pol-II to resume elongation and produce full-length transcripts. Furthermore, we found that in cancer cells, PP2A efficiently dephosphorylates NELF-A, thereby preventing aberrant IEG expression induced by ERK-activating oncogenes. However, when PP2A inhibitor proteins are overexpressed, as is frequently observed in cancers, decreased PP2A activity combined with oncogene-mediated ERK activation conspire to induce NELF-A phosphorylation and IEG upregulation, resulting in tumor progression. Our data delineate previously unexplored roles of ERK and PP2A inhibitor proteins in carcinogenesis.

Elongation factor-specific capture of RNA polymerase II complexes

Transcription of protein-coding genes is regulated by dynamic association of co-factors with RNA polymerase II (RNAPII). The function of these factors and their relationship with RNAPII is often poorly understood. Here, we present an approach for elongation-factor-specific mNET capture (ELCAP) of RNAPII complexes for sequencing and mass spectrometry analysis aimed at investigating the function of such RNAPII regulatory proteins. As proof of principle, we apply ELCAP to the RNAPII-associated proteins SCAF4 and SCAF8, which share an essential role as mRNA anti-terminators but have individual roles at the 3' end of genes. Mass spectrometry analysis shows that both SCAF4 and SCAF8 are part of RNAPII elongation complexes containing 3' end processing factors but depleted of splicing components. Importantly, the ELCAP sequencing (ELCAP-seq) profiles of SCAF4- and SCAF8-RNAPII complexes nicely reflect their function as mRNA-anti-terminators and their competing functions at the end of genes, where they prevent or promote transcriptional readthrough.

MCI-frcnn: A deep learning method for topological micro-domain boundary detection.

Chromatin structural domains, or topologically associated domains (TADs), are a general organizing principle in chromatin biology. RNA polymerase II (RNAPII) mediates multiple chromatin interactive loops, tethering together as RNAPII-associated chromatin interaction domains (RAIDs) to offer a framework for gene regulation. RAID and TAD alterations have been found to be associated with diseases. They can be further dissected as micro-domains (micro-TADs and micro-RAIDs) by clustering single-molecule chromatin-interactive complexes from next-generation three-dimensional (3D) genome techniques, such as ChIA-Drop. Currently, there are few tools available for micro-domain boundary identification. In this work, we developed the MCI-frcnn deep learning method to train a Faster Region-based Convolutional Neural Network (Faster R-CNN) for micro-domain boundary detection. At the training phase in MCI-frcnn, 50 images of RAIDs from Drosophila RNAPII ChIA-Drop data, containing 261 micro-RAIDs with ground truth boundaries, were trained for 7days. Using this well-trained MCI-frcnn, we detected micro-RAID boundaries for the input new images, with a fast speed (5.26 fps), high recognition accuracy (AUROC = 0.85, mAP = 0.69), and high boundary region quantification (genomic IoU = 76%). We further applied MCI-frcnn to detect human micro-TADs boundaries using human GM12878 SPRITE data and obtained a high region quantification score (mean gIoU = 85%). In all, the MCI-frcnn deep learning method which we developed in this work is a general tool for micro-domain boundary detection.