Skip to main content
NCBI home page
PLOS One logoLink to PLOS One
. 2020 Mar 20;15(3):e0223939. doi: 10.1371/journal.pone.0223939

Exploring the overlap between rheumatoid arthritis susceptibility loci and long non-coding RNA annotations

James Ding 1,*, Chenfu Shi 1, John Bowes 1,2, Stephen Eyre 1,2, Gisela Orozco 1,2
Editor: Igor B Rogozin3
PMCID: PMC7083320  PMID: 32196497

Abstract

Whilst susceptibility variants for many complex diseases, such as rheumatoid arthritis (RA), have been well characterised, the mechanism by which risk is mediated is still unclear for many loci. This is especially true for the majority of variants that do not affect protein-coding regions. lncRNA represent a group of molecules that have been shown to be enriched amongst variants associated with RA and other complex diseases, compared to random variants. In order to establish to what degree direct disruption of lncRNA may represent a potential mechanism for mediating RA susceptibility, we chose to further explore this overlap. By testing the ability of annotated features to improve a model of disease susceptibility, we were able to demonstrate a local enrichment of enhancers from immune-relevant cell types amongst RA susceptibility variants (log2 enrichment 3.40). This was not possible for lncRNA annotations in general, however a small, but significant enrichment was observed for immune-enriched lncRNA (log2 enrichment 0.867002). This enrichment was no longer apparent when the model was conditioned on immune-relevant enhancers (log2 enrichment -0.372734), suggesting that direct disruption of lncRNA sequence, independent of enhancer disruption, does not represent a major mechanism by which susceptibility to complex diseases is mediated. Furthermore, we demonstrated that, in keeping with general lncRNA characteristics, immune-enriched lncRNA are expressed at low levels that may not be amenable to functional characterisation.

Introduction

In keeping with other complex diseases, rheumatoid arthritis (RA) susceptibility loci are mainly non-coding, with relatively few variants having a potential impact upon the coding sequence for a protein [1,2]. Enhancers have been identified as likely to mediate disease susceptibility at many loci. Evidence to support this generalization includes demonstrated effects of genome wide association study (GWAS) variants on enhancers at individual loci [3], as well as an enrichment of RA GWAS variants amongst enhancers from relevant cell types [4]. Alternative non-coding elements, such as long non-coding RNA (lncRNA) may also play a role in mediating the increased risk associated with non-coding variants.

lncRNA are a heterogeneous class of molecules that are defined based on a lack of protein-coding potential and a minimum transcribed length of 200 nucleotides. Whilst discrete subcategories exist, such as long intergenic non-coding RNAs, promoter associated-lncRNAs or antisense lncRNA with some discriminatory characteristics, including genomic context, overlapping chromatin marks, length and structure [5], it can still be difficult to discriminate a genuine lncRNA annotation from a product of spurious transcription. Many individual lncRNA have been functionally characterised, with gene regulation featuring frequently amongst the wide variety of roles described. One important subcategory was identified following observations of RNA polymerase II recruitment and transcription at enhancers [6]. Often described as enhancer derived RNA (eRNA), expression of these transcripts is highly correlated with enhancer activity [7] and increasing evidence suggests that these ncRNA may contribute towards enhancer function, although the precise mechanism is still unclear [8].

There is some evidence to suggest that GWAS susceptibility variants are enriched amongst lncRNA [9,10]. However, using conventional methods of determining whether annotations overlap more than can be expected by chance it is difficult to appropriately account for confounding factors, such as chromosomal compartments or chromatin accessibility. Using the ab initio MiTranscriptome assembly (58,648 lncRNA), which was generated using a large collection of RNA sequencing libraries, GWAS SNPs were demonstrated to be enriched in lncRNA, compared to other SNPs tested for in GWAS analyses [9]. This enrichment was also observed using either GWAS SNPs or probabilistically identified causal SNPs (PICS) and lncRNA from the functional annotation of the human genome (FANTOM) cap-analysis gene expression (CAGE) associated transcriptome (CAT) assembly (27,919 lncRNA), generated using CAGE datasets in combination with existing assemblies [10]. Using the tissue specific nature of the FANTOM CAT annotation it was also possible to demonstrate that this enrichment was markedly higher when testing specifically for an enrichment of immune-relevant GWAS PICS in immune-expressed lncRNA transcripts.

Despite these studies, the relevance of lncRNA to the study of individual complex diseases, such as RA, remains unclear. This is especially true given the overlap in genomic locality and function between enhancers and lncRNA. We chose to investigate the overlap between lncRNA annotations, enhancer annotations and GWAS SNPs associated with RA susceptibility (Fig 1), with the aim of establishing to what degree the direct disruption of lncRNA by RA-associated variants may contribute to the mediation of disease risk. Central to our investigation is the use of the fgwas algorithm [11], which tests the ability of individual annotations to improve a probabilistic model of disease susceptibility, constructed using GWAS summary statistics. Using this method a local enrichment is estimated, that takes into consideration the non-random distribution of potentially confounding genomic features. In addition, it is possible to model multiple traits and establish the degree to which they are independently predictive.

Fig 1. Visual interpretation of analysis pipeline.

Fig 1

Open in a new tab

GWAS summary stats were used to inform a probabilistic model of RA susceptibility. Various features were taken from publically available data and their ability to improve this model was tested, features that improve the model can be thought of as enriched amongst RA susceptibility variants.

Materials and methods

Enrichment testing

Enrichment testing was performed using RA summary statistics [1] and fgwas v0.3.6 [11], with the “-cond” option called for conditional analyses. Enrichment estimates are reported as outputted by fgwas, on a log2 scale, such that positive values indicate an enrichment and negative values indicate a depletion of the given annotation amongst disease susceptibility variants. Chromatin state data was obtained from the Roadmap Epigenomics project [12], with the expanded, 18-state model used for all 98 corresponding epigenomes. These chromatin states originate from a hidden Markov model, based on H3K4me3, H3K4me1, H3K36me3, H3K27me3, H3K9me3 and H3K27ac occupancy data. In our analyses, the 18-states were combined to form four exclusive annotations as outlined in Table 1.

Table 1. Consolidation of Roadmap epigenomics 18-state chromatin states into four exclusive annotations.

TSSs Transcription Enhancers Repressed Chromatin
1. Active TSS
2. Flanking TSS
3. Flanking TSS upstream
4. Flanking TSS downstream
14. Bivalent/poised TSS		 5. Strong transcription
6. Weak Transcription		 7. Genic enhancer 1
8. Genic enhancer 2
9. Active enhancer 1
10. Active enhancer 2
11. Weak enhancer
15. Bivalent enhancer		 12. ZNF genes and repeats
13. Heterochromatin
16. Repressed polycomb
17. Weak repressed polycomb
18. Quiescent/low
Open in a new tab

TSS, transcription start site; ZNF, zinc finger. The chromatin state number employed by the Roadmap Epigenomics Project is included to enable easy comparison.

The following lncRNA datasets were interrogated: Lncipedia v5.2 [13], miTranscriptome v2 [9] and FANTOM CAT (robust) [10].Immune-relevant enhancers were defined as genomic regions annotated as enhancers in cell-types originating from “Blood and T cell” or “HSC and B cell” anatomical locations by the Roadmap Epigenomics project [12]. These include a total of 17 partially overlapping primary cell populations, isolated from peripheral and umbilical cord blood; including peripheral blood mononuclear cells, B cells, Natural killer cells, haematopoietic stem cells, and various subtypes of T cells. The definition of Immune-enriched lncRNA is based on the underlying sample ontology and was wholly adopted from Hon et al. requiring: detection in at least 50% of immune-relevant samples, 5 x higher expression in immune-relevant samples than in other samples and P < 0.05 in a one-tailed Mann–Whitney rank sum test [10].

Expression profiling

Raw RNA-seq reads were downloaded from the Roadmap Epigenomics project for primary T-helper cells (Roadmap Epigenomics epigenome ID E043, SRA accession SRR644513 and SRR643766, originating from a 37 y old Hispanic male, and 21 y old Caucasian male, respectively) [12]. Reads were then filtered for quality, adapter content and polyA tails using fastp version 19.7, with default settings and polyX tail trimming enabled. Transcripts were quantified using Salmon version 13.1 [14] using suggested settings (“quant” mode and “–validatemappings”) and using the reference index generated from the FANTOM CAT robust database [10]. Transcripts quantifications (reported as Transcripts per million, TPM) were then remapped to genes and summed for each gene. CAGE transcript counts per million (CPM) were taken from FANTOM CAT for T-helper cells (CL_0000084_T_cell), as published [10].

Statistical difference between the distributions of immune-enriched lncRNA and mRNA abundance was established using a two-sided Z-test, with no assumption of equal variance.

Results

Enhancer annotations from immune-relevant cell types are enriched amongst RA susceptibility variants

An enrichment of RA PICS has previously been demonstrated amongst cis-regulatory elements that are active in T-helper cells and lymphoblastoid cells [4], using data from the high density Immunochip custom SNP array. In order to establish confidence in the ability of fgwas to identify similar enrichments we sought to validate this evidence of enrichment using a more inclusive approach that incorporates the probability of association for all imputed SNPs taken from the most recent European RA GWAS meta-analysis (8 514 610, p value ranges from 1.0 to 1.94 x 10−280) [1]. In order to achieve this, chromatin state data taken from the Roadmap Epigenomics project [12] was incorporated in a model of RA susceptibility. Using enrichment estimates generated using the expanded 18-state model, it is possible to discern an enrichment of certain chromatin states, such as genic enhancers, active enhancers and weak enhancers in immune-relevant cell types, such as B cells, T cells and monocytes (panel A in S1 Fig), however the confidence intervals (CIs) for estimates are broad for many states, likely due to a reduced abundance of these annotations (panel B and C in S1 Fig). The size of CIs was improved by combining states to generate four more easily interpretable annotations (enhancers, transcription start sites (TSSs), transcription, and repressed chromatin; Fig 2). In keeping with our understanding of RA, the highest level of enrichment was observed for enhancer annotations in immune-relevant cell types, with regulatory T cells showing the highest enrichment (log2 enrichment 3.17, 95% CI 2.58; 3.75015). In immune-relevant cell types TSSs were also enriched, whilst repressed chromatin was depleted (Fig 2).

Fig 2. Enrichment of chromatin state groups amongst RA susceptibility variants for 98 cell types.

Fig 2

Open in a new tab

Estimates for enrichment of combined chromatin state groupings are illustrated for all 98 cell types annotated within the Roadmap Epigenomics 18-state model. Cell-types are ordered and coloured according to the clustering established by the Roadmap Epigenomics project, with immune-relevant cell types coloured green. Estimates and confidence intervals are clipped at axis limits, where applicable.

lncRNA annotations show negligible enrichment amongst RA susceptibility variants

We applied fgwas to test for an enrichment of lncRNA amongst RA susceptibility variants, using lncipedia, a large lncRNA database curated from a number of sources [13], the MiTranscriptome assembly [9], and the FANTOM CAT assembly [10]. Using fgwas, MiTranscriptome lncRNA genes show a level of depletion amongst RA susceptibility variants comparable to repressed chromatin. MiTranscriptome lncRNA exons and both genes and exons from either FANTOM CAT or lncipedia all show negligible enrichment (Fig 3A).

Fig 3. Enrichment of lncRNA annotations amongst RA susceptibility variants.

Fig 3

Open in a new tab

Estimates for the enrichment of genic (black circle) and exonic (grey diamond) annotations from a variety of lncRNA containing databases, including 95% confidence intervals (A). Separate estimates are included for annotations identified as exhibiting enriched expression in immune-relevant cells (B).

lncRNA annotations with enriched expression in immune-relevant cells are subtly enriched amongst RA susceptibility variants

Uniquely, FANTOM CAT transcripts are associated with tissue specific expression data. As in the original FANTOM CAT publication, we took advantage of this additional information, to test for an enrichment of lncRNA whose expression is enriched in immune-relevant cell types amongst RA susceptibility loci. This approach demonstrated a subtle enrichment of lncRNA genes (log2 enrichment 0.867, 95% CI 0.0554; 1.57) and similar level of enrichment for their exons, albeit with an increased confidence interval (log2 enrichment 0.799, 95% CI 2.30; 1.94). FANTOM CAT mRNA annotations, whose expression is enriched in immune-relevant cell types, were included in order to provide a comparison. Genic mRNA annotations showed a similar level of enrichment as lncRNA genes, with mRNA exons exhibiting slightly higher enrichment (Fig 3B).

The subtle enrichment of immune-enriched lncRNA observed is not independent of immune-relevant enhancer annotations

Given the strong enrichment of immune-relevant enhancers amongst RA susceptibility loci and the established overlap between lncRNA annotations and enhancers, we were interested to investigate the independence of these variables using fgwas. In this conditional analysis, a residual enrichment of immune-enriched FANTOM CAT annotations was tested after the enrichment of immune-relevant enhancers (log2 enrichment 3.40, 95% CI 2.54; 4.58) was accounted for (Fig 4). Interestingly, both lncRNA and mRNA annotations no longer show significant enrichment, indicating that once enrichment of susceptibility variants in enhancers has been accounted for no remaining enrichment of mRNA or lncRNA is apparent.

Fig 4. Enrichment of immune-enriched lncRNA amongst RA susceptibility variants after conditioning on chromatin state data.

Fig 4

Open in a new tab

The influence of immune-relevant enhancers (red circle) was fixed in a probabilistic model of RA susceptibility to determine whether the subtle enrichment of FANTOM CAT immune-enriched lncRNA or mRNA adds any additional predictive information and is therefore independently enriched. Genic (black circle) and exonic (grey diamond) annotations were both tested. As may be expected given the magnitude of enrichments observed, after accounting for the effect of immune-enriched FANTOM CAT annotations, the residual enrichment of immune-relevant enhancers is not dramatically reduced (S2 Fig).

Immune-enriched lncRNA are expressed at low levels in RA relevant cell types

lncRNA are generally considered to exhibit low expression levels and high tissue specificity that can make them difficult to study using conventional methods. Given their enrichment for RA susceptibility variants we were interested to establish whether this description applied to FANTOM CAT immune-enriched lncRNA in an RA relevant cell type. Using randomly primed Roadmap Epigenomics RNA-seq data from primary T-helper cells, the distribution of expression levels for FANTOM CAT immune-enriched lncRNA is significantly lower than that of FANTOM CAT immune-enriched mRNA (p = 1.04 x 10−34, Fig 5A median lncRNA transcripts per million reads (TPM); 0.257, vs 124 for mRNA). The same is true using less-conventional expression profiling methods, such as those employed by FANTOM CAT, which offer improved sensitivity for the detection of transcripts of low abundance (p = 1.01 x 10−24, Fig 5B median lncRNA counts per million reads (CPM); 1.43, vs 71.7 for mRNA). 90% of immune-enriched lncRNA have abundance lower than 85.2% of immune-enriched mRNA in Roadmap Epigenomics RNA-seq data or 75.6% of immune-enriched mRNA in FANTOM CAT CAGE data.

Fig 5. Distribution of immune-enriched lncRNA and mRNA expression levels in primary T-helper cells.

Fig 5

Open in a new tab

Staggered bars are used to illustrate the proportion of transcripts whose expression falls in bins of 25 million transcripts, or counts, in Roadmap Epigenomics RNA-seq data (A) and FANTOM CAT CAGE data (B), respectively.

Discussion

By incorporating both cell-type specific enhancer and lncRNA annotations into a probabilistic model of RA susceptibility it is possible to demonstrate their respective levels of enrichment amongst RA susceptibility variants. Whilst previous studies have demonstrated an enrichment of lncRNA compared with randomly shuffled annotations these analyses fail to take into consideration the complex organisation of the genome and are easily confounded by alternative features. In our analysis, which incorporated lncRNA from various databases, it was only possible to demonstrate a subtle enrichment of lncRNA whose expression was previously identified as being enriched in relevant cell-types.

By conditioning a model of RA susceptibility on enhancer annotations from immune-relevant cell types it is possible to test the independence of additional features. This demonstrated that the subtle enrichment observed for immune-relevant lncRNA is entirely explained by immune-relevant enhancer annotations. Interestingly, the same is true of mRNA annotations, indicating that in both instances the primary influence of susceptibility variants is in affecting non-coding regulatory elements, with any effect on mRNA and lncRNA being secondary and/or indirect. This suggests that the majority of genetic variance in disease susceptibility is mediated through disruption of regulatory elements and not through direct disruption of transcript sequences. This observation is in keeping with those made previously, relating to the minimal overlap between the coding sequence of mRNA and RA susceptibility variants. It is also consistent with the existence of well-characterised effects of RA susceptibility variants on coding regions, such as for the HLA proteins [15], which convey a significant, but not exhaustive proportion of risk.

Similarly, this analysis does not rule out the relevance of lncRNA at individual loci, in fact it is worth noting that there is evidence to suggest that C5T1lncRNA may mediate risk at a RA risk locus located at chromosomal position 9q33.2, with RA associated variants falling within a C5T1lncRNA exonic region [16]. Our analysis, however, precludes a general effect of RA susceptibility variants on the transcribed sequence of lncRNA independent of enhancer disruption. Sequence-specific functions of lncRNA are, therefore, unlikely to mediate a significant proportion of the risk-modifying effect associated with RA susceptibility variants. Furthermore, the distribution of FANTOM-CAT immune-enriched lncRNA expression levels in primary T-helper cells highlight the difficulties associated with studying lncRNA, whose expression is typically very low and highly cell-type specific. It should however be highlighted that whilst these obstacles may complicate the functional characterisation of lncRNA, low, highly tissue specific expression on its own does not preclude a biologically significant role.

In our analysis, we have sought to highlight the difficulties associated with characterising a role for lncRNA that overlap disease susceptibility variants in mediating disease susceptibility. These difficulties have also undoubtedly contributed to limitations in the accuracy and completeness of the lncRNA annotations used, where lncRNA with low expression or with expression levels that are highly specific to a certain developmental stage, condition, or cell type are likely to be underrepresented.

Many mechanisms have been described, whereby lncRNA may impact disease susceptibility irrespective of disruptions to their DNA sequence and our findings do not limit the likelihood that these mechanisms may be important in RA susceptibility. For example, transcription of a lncRNA at a particular locus may reinforce a permissive chromatin state that is required for accurate regulation of another gene. However, where such a mechanism is found to be important in mediating the effects of a disease susceptibility variant, it is perhaps most likely that the primary impact of the disease susceptibility variants is in disrupting a regulatory element that controls the lncRNA expression. In such instances, it is inherently difficult to disentangle the functional contribution of the lncRNA and regulatory element.

The analyses performed are specific to RA, however it is assumed that similar results would be reached using GWAS data, enhancer annotations and lncRNA annotations relevant to other diseases. This would, therefore, suggest that the enrichment of lncRNA annotations amongst GWAS variants observed by others [9,10], may result from a high degree of overlap between regulatory features such as enhancers and lncRNA and other confounding features, such as active and inactive genomic compartments. As a generalisation, when it comes to functional characterization of variants associated with complex genetic disorders, sequence-specific lncRNA functions is unlikely to represent an attractive area for study; lncRNA are not independently enriched amongst such variants and are difficult to study, due to low expression levels. Despite these results, dysregulation of lncRNA expression could still play a role in RA and similar diseases, with disease associated variants affecting regulatory elements, such as enhancers that control lncRNA expression.

fgwas represents a useful tool for studying the enrichment of different features amongst susceptibility variants, especially when used in combination with Roadmap Epigenomics data in order to identify cell-types and tissues that are relevant for disease susceptibility. Whilst using this tool we observed that the number and size of annotations that are tested have a strong influence, both on the confidence with which any enrichment is estimated, as well as on the extent of that enrichment. It is likely that this may explain some of the subtle differences observed between different lncRNA databases.

Our analyses highlight the caveats associated with inferring functional relevance for a given feature, based purely on the observation of enrichment over a genomic background, as well as the care that must be taken when attempting to interpret such enrichments. By deriving enhancer and lncRNA annotations from entirely different sources we have tried to ensure that the demonstrated dependence is not self-fulfilling, as it may have been if we defined immune-relevant lncRNA based on underlying chromatin states.

In conclusion, using fgwas and Roadmap Epigenomics chromatin state data it is possible to identify cell types and chromatin states of relevance to complex diseases, such as RA. In the case of RA this is predominantly enhancers and transcription start sites from immune-relevant cell types. It is also possible to test the association of alternative features and establish their independence from chromatin states. Here, a previously described enrichment of lncRNA amongst GWAS susceptibility loci was explored for RA. Immune-enriched lncRNA from the FANTOM-CAT database were found to be enriched amongst RA susceptibility loci, however, this enrichment was not apparent when chromatin-state data was taken into account.

Our results suggest that regulatory elements, such as enhancers, are likely to mediate the vast majority of variance in risk associated with RA and other complex diseases, with no substantial independent contribution being made by direct disruption of lncRNA sequences. Because of this, and the difficulties associated with detecting transcripts of such low abundance, sequence-specific lncRNA function does not represent the most attractive area for study with respect to RA susceptibility, except in the case of in depth characterisation of individual loci.

Supporting information

S1 Fig. Enrichment of chromatin state annotations amongst RA susceptibility variants.

Estimates for enrichment of individual states are illustrated for 98 cell types using the Roadmap Epigenomics 18-state model (a). Similar states were grouped into four groups for all immune-relevant primary cell-types (b), as individual states often gave very broad 95% confidence intervals (c). Cell-types are ordered according to the clustering established by the Roadmap Epigenomics project, with chromatin states reordered according to their subsequent grouping. Estimates and confidence intervals are clipped at axis limits, where applicable.

(TIF)

Click here for additional data file. (2.3MB, tif)
S2 Fig. Enrichment of immune-relevant enhancers amongst RA susceptibility variants after conditioning on immune-enriched transcripts.

The influence of FANTOM CAT immune-enriched lncRNA and mRNA was fixed in a probabilistic model of RA susceptibility to confirm the independent enrichment of immune-relevant enhancer chromatin states.

(TIF)

Click here for additional data file. (301.1KB, tif)

Acknowledgments

The authors would like to acknowledge the assistance given by IT Services and the use of the Computational Shared Facility at The University of Manchester.

Data Availability

The datasets analysed during the current study are available from FANTOM, GWAS Catalogue, LNCipedia, MiTranscriptome, Roadmap Epigenomics project, and Sequenced Read Archive data repositories: http://fantom.gsc.riken.jp/5/suppl/Hon_et_al_2016/data/, ftp://ftp.ebi.ac.uk/pub/databases/gwas/summary_statistics/OkadaY_24390342_GCST002318, https://lncipedia.org/download, http://mitranscriptome.org/download/, https://egg2.wustl.edu/roadmap/data/byFileType/chromhmmSegmentations/ChmmModels/core_K27ac/jointModel/final/, and https://www.ncbi.nlm.nih.gov/sra, respectively.

Funding Statement

This research was funded by the Wellcome Trust and Versus Arthritis and supported by the NIHR Manchester Biomedical Research Centre. JD, CS and GO are funded by the Wellcome Trust (JD and GO, 207491/Z/17/Z; CS, 215207/Z/19/Z) and by Versus Arthritis (21754 and 21348). JB and SE are funded by Versus Arthritis (JB, 21754; SE, 21754 and 21348). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. There was no additional external funding received for this study.

References

  • 1.Okada Y, Wu D, Trynka G, Raj T, Terao C, Ikari K, et al. Genetics of rheumatoid arthritis contributes to biology and drug discovery. Nature. 2013;506: 376–381. 10.1038/nature12873 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Eyre S, Bowes J, Diogo D, Lee A, Barton A, Martin P, et al. High-density genetic mapping identifies new susceptibility loci for rheumatoid arthritis. Nat Genet. 2012;44: 1336–40. 10.1038/ng.2462 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Simeonov DR, Gowen BG, Boontanrart M, Roth TL, Gagnon JD, Mumbach MR, et al. Discovery of stimulation-responsive immune enhancers with CRISPR activation. Nature. 2017;549: 111–115. 10.1038/nature23875 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Farh KK-H, Marson A, Zhu J, Kleinewietfeld M, Housley WJ, Beik S, et al. Genetic and epigenetic fine mapping of causal autoimmune disease variants. Nature. 2014;518: 337–43. 10.1038/nature13835 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Jarroux J, Morillon A, Pinskaya M. History, discovery, and classification of lncRNAs Advances in Experimental Medicine and Biology. Springer, Singapore; 2017. pp. 1–46. 10.1007/978-981-10-5203-3_1 [DOI] [PubMed] [Google Scholar]
  • 6.Kim T-K, Hemberg M, Gray JM, Costa AM, Bear DM, Wu J, et al. Widespread transcription at neuronal activity-regulated enhancers. Nature. 2010;465: 182–7. 10.1038/nature09033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Andersson R, Gebhard C, Miguel-Escalada I, Hoof I, Bornholdt J, Boyd M, et al. An atlas of active enhancers across human cell types and tissues. Nature. 2014;507: 455–61. 10.1038/nature12787 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Lam MTY, Li W, Rosenfeld MG, Glass CK. Enhancer RNAs and regulated transcriptional programs. Trends Biochem Sci. 2014;39: 170–82. 10.1016/j.tibs.2014.02.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Iyer MK, Niknafs YS, Malik R, Singhal U, Sahu A, Hosono Y, et al. The landscape of long noncoding RNAs in the human transcriptome. Nat Genet. 2015;47: 199–208. 10.1038/ng.3192 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Hon C-C, Ramilowski JA, Harshbarger J, Bertin N, Rackham OJL, Gough J, et al. An atlas of human long non-coding RNAs with accurate 5′ ends. Nature. 2017;543: 199–204. 10.1038/nature21374 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Pickrell JK. Joint analysis of functional genomic data and genome-wide association studies of 18 human traits. Am J Hum Genet. 2014;94: 559–73. 10.1016/j.ajhg.2014.03.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Kundaje A, Meuleman W, Ernst J, Bilenky M, Yen A, Heravi-Moussavi A, et al. Integrative analysis of 111 reference human epigenomes. Nature. 2015;518: 317–330. 10.1038/nature14248 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Volders P-J, Anckaert J, Verheggen K, Nuytens J, Martens L, Mestdagh P, et al. LNCipedia 5: towards a reference set of human long non-coding RNAs. Nucleic Acids Res. 2019;47: D135–D139. 10.1093/nar/gky1031 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Patro R, Duggal G, Love MI, Irizarry RA, Kingsford C. Salmon provides fast and bias-aware quantification of transcript expression. Nat Methods. 2017;14: 417–419. 10.1038/nmeth.4197 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Raychaudhuri S, Sandor C, Stahl EA, Freudenberg J, Lee H-S, Jia X, et al. Five amino acids in three HLA proteins explain most of the association between MHC and seropositive rheumatoid arthritis. Nat Genet. 2012;44: 291–6. 10.1038/ng.1076 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Messemaker TC, Frank-Bertoncelj M, Marques RB, Adriaans A, Bakker AM, Daha N, et al. A novel long non-coding RNA in the rheumatoid arthritis risk locus TRAF1-C5 influences C5 mRNA levels. Genes Immun. 2015; 10.1038/gene.2015.54 [DOI] [PubMed] [Google Scholar]

Decision Letter 0

Igor B Rogozin

17 Dec 2019

PONE-D-19-27368

Exploring the overlap between rheumatoid arthritis susceptibility loci and long non-coding RNA annotations

PLOS ONE

Dear Dr Ding,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please, implement suggestions, I do not see major problems.

We would appreciate receiving your revised manuscript by Jan 31 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

  • A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Igor B. Rogozin

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements:

  1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.plosone.org/attachments/PLOSOne_formatting_sample_main_body.pdf and http://www.plosone.org/attachments/PLOSOne_formatting_sample_title_authors_affiliations.pdf

  2.  

    Thank you for stating in your Funding Statement:

     [The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.]. 

    Please provide an amended statement that declares *all* the funding or sources of support (whether external or internal to your organization) received during this study, as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now.  Please also include the statement “There was no additional external funding received for this study.” in your updated Funding Statement.

    Please include your amended Funding Statement within your cover letter. We will change the online submission form on your behalf.

Additional Editor Comments (if provided):

Dear Dr. Ding,

Please, implement suggestions, I do not see major problems.

Best regards, Igor

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Ding and colleagues set out to discern whether the direct disruption of long non-coding RNA (lncRNA) could be a mechanism involve in the association of SNPs and rheumatoid arthritis (RA) susceptibility. They assessed the question by evaluating the overlap between SNPs associated with RA susceptibility, lncRNA annotations and enhancer annotations. The authors concluded that the RA susceptibility genetic variants are more likely to act via enhancers than lncRNA. Since the subtle enriched overlap between SNPs associated with RA and lncRNAs are not independent from the strong overlapped enrichment between RA SNPs and enhancers.

This study contributes to comprehend which are the mechanisms behind the statistical associations of SNPs with the risk to develop RA. Importantly, the authors present an improvement on the methodology to assess functional genomic information in relation to GWAS results. This is to condition the analysis with the chromatin state information, in order to avoid cofounders on the enrichment analysis.

I would like to point minor changes which may improve the future reproducibility of the study and a couple of challenges that may or may not be in the scope of the study.

1. Would be possible for the authors to create a graph or a visual interpretation of the analysis pipeline implemented in the study?

2. Please add more details about the RA GWAS summary statistics employ in the analysis. For instance, how many SNPs and max and min p-value of association (~range of p-values). In order to provide the reader relevant information without having to search for the referenced paper.

3. Could you please add more details in the methodology to the “immune-relevant enhancers definition” (page 6 lines 98 to 101)? For example, the flow cytometry antibodies (or markers) reference by the Roadmap Epigenome Project to the “Blood and T cells” or “HSC and B cell”. It could be done adding them to the text or a link to Roadmap Epigenome Project.

4. A similar query as the previous one for the expression profiling section. Please add the specific cell type to which the SRA accession numbers refers to. For instance, is it from adult or fetal cells? Which flow cytometry markers?

5. Could the authors briefly explain in the methods section how to interpret the statistic (i.e. log2 enrichment) they used to “quantify” enrichment? E.g. how does it relate to the two-side Welch’s test or is it the output of the fgwas? This to facilitate the understanding added to the descriptors of the enrichment measure (i.e. negligible in line 159)

Other points

1. Did the authors consider to test the lncRNA/enhancer enrichment in other cellular types known to be relevant in the inflammation site of RA, such fibroblasts?

2. Would it be possible to narrow down the immune-related enhancer enrichment to more specific cell types? In a similar fashion as presented in figure 2b.

3. Unfortunately, the figure 1 was not readable, possibly due to a combination of the original format and the journal submission system. Therefore, my comments are based on the rest of the content of the manuscript.

Reviewer #2: This manuscript reports an analysis of the potential role of genetic variants, identified from genome-wide association studies (GWAS) in rheumatoid arthritis, on mediating their effects through long non-coding RNAs (lncRNAs). The authors use publically available GWAS data, chromatin and RNA-seq data from the Roadmap Epigenomics Project and the fgwas algorithm to undertake these analyses.

The authors report a significant enrichment for immune-enriched lncRNA, but this enrichment was no longer apparent when conditioned on immune-relevant enhancers. The authors therefore suggest that direct disruption of lncRNA sequence, independent of enhancer disruption, does not represent a major mechanism by which susceptibility to complex diseases is mediated.

This paper is clearly written and well executed. I have some generally minor comments; these relate primarily to a more nuanced and more balanced discussion of the data.

Materials and Methods

For clarity and ease of reading, it would be better to summarise the data on the Roadmap Epigenomics 18-state model in a table.

Results

Lines 205-212. The assumption of normality for the two-sided Welch’s t-test to test “Statistical difference between the distributions of immune-enriched lncRNA and mRNA abundance” is clearly not valid (Fig 4). The authors should either use a more appropriate alternative test, or comment on this.

Discussion

I would like to see included some more nuanced and balanced discussion relating to the effects of lncRNAs that are not mediated through direct disruption of DNA sequence. This should include some discussion of the fact that the relatively low lncRNA expression levels may not correlate with effect size /biological relevance of specific lncRNAs. lncRNAs also show higher developmental stage specificity (as well as cell type specificity) than mRNAs, with regard to the data sources used, and the fact that these publically available resources reflect the healthy, rather than disease, state.

Minor comments on syntax

Methods. Enrichment testing. Define ‘TSS’ and ‘HSC’ at first use.

Results, line 186. Fig 3 is duplicated.

Typo fig 4A axis label.

Line 204. Complexes diseases.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2020 Mar 20;15(3):e0223939. doi: 10.1371/journal.pone.0223939.r002

Author response to Decision Letter 0

31 Jan 2020

Comments from all reviewers are gratefully received; we believe that by implementing the following recommendations the rigor and ease of understanding of our manuscript has been improved and would like to thank all reviewers for their contributions towards this.

A full list of how each of the reviewers’ comments have been addressed is found below. Please note that line numbers relate to the revised manuscript with track changes.

I hope that you are satisfied with the manner in which these comments have been addressed and agree that they have improved the manuscript. Thank you once again for contributing to our manuscript in such a constructive manner.

Yours sincerely,

Dr James Ding

REVIEWER #1:

1. Would be possible for the authors to create a graph or a visual interpretation of the analysis pipeline implemented in the study?

A visual interpretation of the analysis pipeline has been added as an additional figure (Fig1) and referenced in the text at the end of the introduction. We hope this clarifies the approach taken.

2. Please add more details about the RA GWAS summary statistics employ in the analysis. For instance, how many SNPs and max and min p-value of association (~range of p-values). In order to provide the reader relevant information without having to search for the referenced paper.

We thank the reviewers for this comment as we had neglected to highlight that we restricted our analysis to the European meta-analysis as we believe this corresponds best to the ethnicity of the donors that contribute to other data-sources in our analysis. We apologise for this oversight and have sought to clarify this in the text, such that line 160-162 now reads: “…using a more inclusive approach that incorporates the probability of association for all imputed SNPs taken from the most recent European RA GWAS meta-analysis (8 514 610 SNPs, p value ranges from 1.0 to 1.94 x 10-280) [1]”. We hope that this clarifies that no threshold was used to determine which SNPs contributed to our analysis.

3. Could you please add more details in the methodology to the “immune-relevant enhancers definition” (page 6 lines 98 to 101)? For example, the flow cytometry antibodies (or markers) reference by the Roadmap Epigenome Project to the “Blood and T cells” or “HSC and B cell”. It could be done adding them to the text or a link to Roadmap Epigenome Project.

We have included additional detail in the methods section of our paper to describe how immune-relevant enhancers were defined. This includes reference to the 6 histone marks used to contribute to the Roadmap Epigenomics Projects 18-state model (lines 109-111) and additional description of the cell types contributing to the Roadmap Epigenomics Project “Blood and T cells” or “HSC and B cell” cell type/tissue group. The relevant section (lines 125-130) now reads: “Immune-relevant enhancers were defined as genomic regions annotated as enhancers in cell-types originating from “Blood and T cell” or “HSC and B cell” anatomical locations by the Roadmap Epigenomics project [12]. These include a total of 17 partially overlapping primary cell populations, isolated from peripheral and umbilical cord blood; including peripheral blood mononuclear cells, B cells, Natural killer cells, haematopoietic stem cells, and various subtypes of T cells.”

Additional detail can be found in Figure 2 and Supplementary Table 1 of Kundaje et al. 2015 (referenced in our manuscript as reference 12). In addition, experimental protocols from the roadmap epigenomics website detail that “Hematopoietic cells were provided as a service by the S. Heimfeld Laboratory at the Fred Hutchinson Cancer Research Center, Seattle, WA and include CD3+, CD4+, CD8+, CD14+, CD19+/CD20+, CD34+, and CD56+ cells, from both mobilized and non-mobilized donors. Cells were obtained from human leukapheresis product using standard procedures. Briefly, the lymphocyte subclasses were isolated by immunomagnetic separation using the CliniMACS affinity-based technology (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany) according to the manufacturer’s recommendations.”

4. A similar query as the previous one for the expression profiling section. Please add the specific cell type to which the SRA accession numbers refers to. For instance, is it from adult or fetal cells? Which flow cytometry markers?

We have included additional detail taken from the National Centre for Biotechnology Information’s database, corresponding to the age, gender and ethnicity of these samples (lines 137-139). Unfortunately it has not been possible to include flow-cytometry/immunomagnetic markers as these are not clearly reported by the Roadmap Epigenomics Project. In Supplementary Table 1 of Kundaje et al. 2015 these cells are referred to as CD4+ and CD25- “Primary T helper cells from peripheral blood”.

5. Could the authors briefly explain in the methods section how to interpret the statistic (i.e. log2 enrichment) they used to “quantify” enrichment? E.g. how does it relate to the two-side Welch’s test or is it the output of the fgwas? This to facilitate the understanding added to the descriptors of the enrichment measure (i.e. negligible in line 159)

This has been clarified, through the inclusion of an additional sentence in the methods section (line 104-107): “Enrichment estimates are reported as outputted by fgwas, on a log2 scale, such that positive values indicate an enrichment and negative values indicate a depletion of the given annotation amongst disease susceptibility variants.”

Other points

1. Did the authors consider to test the lncRNA/enhancer enrichment in other cellular types known to be relevant in the inflammation site of RA, such fibroblasts?

We agree that it would be of benefit to the manuscript to include additional disease relevant cell types, such as synovial fibroblasts, or fibroblast-like synoviocytes (FLS), however data sufficient to perform a comparable analysis with that performed for other cell types is not publically available. FLS show many characteristics common with fibroblasts, but they also secrete unique proteins, that are normally absent in other fibroblast lineages. Nevertheless, two primary foreskin fibroblast cell types are included in what is now Fig2, corresponding to Roadmap Epigenomics epigenome IDs E055 and E056. An enrichment of enhancer annotations from these samples was not observed.

2. Would it be possible to narrow down the immune-related enhancer enrichment to more specific cell types? In a similar fashion as presented in figure 2b.

This has effectively been performed in what is now Fig2, which shows the extent to which enhancers annotated in all Roadmap Epigenomics epigenomes are enriched amongst RA susceptibility variants. Whilst there is some variability in the extent of this enrichment, and we report the cell type which gives the highest enrichment estimate (regulatory T cells, line 174), confidence intervals overlap to the extent that no firm comparisons can be drawn.

3. Unfortunately, the figure 1 was not readable, possibly due to a combination of the original format and the journal submission system. Therefore, my comments are based on the rest of the content of the manuscript.

This is regrettable, especially since it appears to have relevance to the two previous points. We apologise for any contribution we have made to this and would encourage this reviewer to review this figure on bioRxiv, should it remain unreadable upon submission of our revised manuscript (https://doi.org/10.1101/804880).

REVIEWER #2:

1. Materials and Methods: For clarity and ease of reading, it would be better to summarise the data on the Roadmap Epigenomics 18-state model in a table.

This information has been transferred to a table, which we agree aids the clarity and ease of reading.

2. Results: Lines 205-212. The assumption of normality for the two-sided Welch’s t-test to test “Statistical difference between the distributions of immune-enriched lncRNA and mRNA abundance” is clearly not valid (Fig 4). The authors should either use a more appropriate alternative test, or comment on this.

Thank you for this comment, we had struggled to calculate the statistical significance of the difference between the distributions shown in what is now Fig 5, but have now found a more suitable test. Due to the large sample sizes and the central limit theorem the Z-score can be assumed to approximate a normal distribution. The significance of both comparisons is now reported as “p = 1.04 x 10-34” and “p = 1.01 x 10-24” in the results section, with the methods section (lines 147-149) reading: “Statistical difference between the distributions of immune-enriched lncRNA and mRNA abundance was established using a two-sided Z-test, without assuming equal variance.”

3. Discussion: I would like to see included some more nuanced and balanced discussion relating to the effects of lncRNAs that are not mediated through direct disruption of DNA sequence. This should include some discussion of the fact that the relatively low lncRNA expression levels may not correlate with effect size /biological relevance of specific lncRNAs. lncRNAs also show higher developmental stage specificity (as well as cell type specificity) than mRNAs, with regard to the data sources used, and the fact that these publically available resources reflect the healthy, rather than disease, state.

We have expanded our discussion (lines 289-307) in order to better cover lncRNA that are either not captured within the data sources used or function independent of their underlying DNA sequence. This includes highlighting the potential disparity between the extent to which a lncRNA may be expressed and its biological significance.

Minor comments on syntax

1. Methods. Enrichment testing. Define ‘TSS’ and ‘HSC’ at first use.

2. Results, line 186. Fig 3 is duplicated.

3. Typo fig 4A axis label.

4. Line 204. Complexes diseases.

These minor comments have all been acted upon with the underlying errors having been corrected. Thank you for your careful revision.

Attachment

Submitted filename: Response to Reviewers.docx

Click here for additional data file. (35.7KB, docx)

Decision Letter 1

Igor B Rogozin

20 Feb 2020

Exploring the overlap between rheumatoid arthritis susceptibility loci and long non-coding RNA annotations

PONE-D-19-27368R1

Dear Dr. Ding,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Igor B. Rogozin

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

The authors implemented suggestions, the paper is acceptable.

Reviewers' comments:

Acceptance letter

Igor B Rogozin

4 Mar 2020

PONE-D-19-27368R1

Exploring the overlap between rheumatoid arthritis susceptibility loci and long non-coding RNA annotations

Dear Dr. Ding:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Igor B. Rogozin

Academic Editor

PLOS ONE

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    S1 Fig. Enrichment of chromatin state annotations amongst RA susceptibility variants.

    Estimates for enrichment of individual states are illustrated for 98 cell types using the Roadmap Epigenomics 18-state model (a). Similar states were grouped into four groups for all immune-relevant primary cell-types (b), as individual states often gave very broad 95% confidence intervals (c). Cell-types are ordered according to the clustering established by the Roadmap Epigenomics project, with chromatin states reordered according to their subsequent grouping. Estimates and confidence intervals are clipped at axis limits, where applicable.

    (TIF)

    Click here for additional data file. (2.3MB, tif)
    S2 Fig. Enrichment of immune-relevant enhancers amongst RA susceptibility variants after conditioning on immune-enriched transcripts.

    The influence of FANTOM CAT immune-enriched lncRNA and mRNA was fixed in a probabilistic model of RA susceptibility to confirm the independent enrichment of immune-relevant enhancer chromatin states.

    (TIF)

    Click here for additional data file. (301.1KB, tif)
    Attachment

    Submitted filename: Response to Reviewers.docx

    Click here for additional data file. (35.7KB, docx)

    Data Availability Statement

    The datasets analysed during the current study are available from FANTOM, GWAS Catalogue, LNCipedia, MiTranscriptome, Roadmap Epigenomics project, and Sequenced Read Archive data repositories: http://fantom.gsc.riken.jp/5/suppl/Hon_et_al_2016/data/, ftp://ftp.ebi.ac.uk/pub/databases/gwas/summary_statistics/OkadaY_24390342_GCST002318, https://lncipedia.org/download, http://mitranscriptome.org/download/, https://egg2.wustl.edu/roadmap/data/byFileType/chromhmmSegmentations/ChmmModels/core_K27ac/jointModel/final/, and https://www.ncbi.nlm.nih.gov/sra, respectively.

    Articles from PLoS ONE are provided here courtesy of PLOS

    RESOURCES