Statistically based splicing detection reveals neural enrichment and tissue-specific induction of circular RNA during human fetal development.

Szabo L, Morey R, Palpant NJ, Wang PL, Afari N, Jiang C, Parast MM, Murry CE, Laurent LC, Salzman J. · 2015

Read Statistically based splicing detection reveals neural enrichment and tissue-specific induction of circular RNA during human fetal development. by Szabo L, Morey R, Palpant NJ, Wang PL, Afari N, Jiang C, Parast MM, Murry CE, Laurent LC, Salzman J. (2015) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc4506483.txt.

Introduction

Linda Szabo and nine co-authors present a new way to find circular RNA in a 2015 Genome Biology paper. Circular RNAs are RNA loops made when a gene's parts are spliced in a ring. They turn up across many life forms, yet what most of them do is still unclear. Finding them with confidence is the first problem, since weak counts can mislead.

The group's method judges splicing events with statistics rather than raw read counts, and it can use both known and unmarked splice sites. Applied to human fetal tissue, it showed broad rises in circular RNA as organs form, with tissue-specific patterns in heart and lung. Brain areas such as the frontal cortex stood out, rich in genes where the circular form is the main product.

Catalog record

CREATORSzabo L, Morey R, Palpant NJ, Wang PL, Afari N, Jiang C, Parast MM, Murry CE, Laurent LC, Salzman J.
YEAR2015
COLLECTIONPAPERS
FORMATTXT
FILE SIZE125 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC4506483

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Statistically based splicing detection reveals neural enrichment and tissue-specific induction of circular RNA during human fetal development By Linda Szabo, Robert Morey, Nathan J Palpant, Peter L Wang, Nastaran Afari, Chuan Jiang, Mana M Parast, Charles E Murry, Louise C Laurent, Julia Salzman Published in: Genome Biology DOI: 10.1186/s13059-015-0690-5 ABSTRACT Abstract Background The pervasive expression of circular RNA is a recently discovered feature of gene expression in highly diverged eukaryotes, but the functions of most circular RNAs are still unknown. Computational methods to discover and quantify circular RNA are essential. Moreover, discovering biological contexts where circular RNAs are regulated will shed light on potential functional roles they may play. Results We present a new algorithm that increases the sensitivity and specificity of circular RNA detection by discovering and quantifying circular and linear RNA splicing events at both annotated and un-annotated exon boundaries, including intergenic regions of the genome, with high statistical confidence. Unlike approaches that rely on read count and exon homology to determine confidence in prediction of circular RNA expression, our algorithm uses a statistical approach. Using our algorithm, we unveiled striking induction of general and tissue-specific circular RNAs, including in the heart and lung, during human fetal development. We discover regions of the human fetal brain, such as the frontal cortex, with marked enrichment for genes where circular RNA isoforms are dominant. Conclusions The vast majority of circular RNA production occurs at major spliceosome splice sites; however, we find the first examples of developmentally induced circular RNAs processed by the minor spliceosome, and an enriched propensity of minor spliceosome donors to splice into circular RNA at un-annotated, rather than annotated, exons. Together, these results suggest a potentially significant role for circular RNA in human development. Electronic supplementary material The online version of this article (doi:10.1186/s13059-015-0690-5) contains supplementary material, which is available to authorized users. BACKGROUND The pervasive expression of circular RNA from protein- and non-coding loci is a recently discovered feature of highly diverged eukaryotic gene expression programs, conserved from humans to very simple organisms such as fungi [1–5]. Isolated reports of expression of circular RNAs from single genes have existed for decades, but mainly due to technological and methodological biases were generally considered to be rare splicing “mistakes” until quite recently [1]. We and others have shown that, in humans, thousands of genes have circular RNA isoforms, their expression relative to that of cognate linear RNA and their alternative splicing varying by cell type [3, 6]. For hundreds of genes, the circular RNA isoform is more abundant than linear RNA from the same locus, raising the intriguing possibility of functional roles for these molecules [1, 3, 6]. While isolated examples of circular RNA acting as microRNA sponges have been reported [2, 7], comprehensive detection and quantification of circular RNA is a necessary foundation for future studies aimed at discovery of additional circular RNA functions and elucidation of mechanisms for circular RNA regulation. Identification of biological systems in which the expression of circular RNA differs according to time, space, or cell type may provide insights into both the function and regulation of circular RNA. In order to screen large numbers of diverse datasets for this purpose, precise statistical algorithms to quantify circular and linear RNA splicing are required, and currently available algorithms for doing so have significant shortcomings. Lack of rigorous statistical testing and/or biases in ascertainment of circular RNA expression have the potential to reduce both the sensitivity and the accuracy of circular RNA detection and quantification, and can in fact lead to artifactual discovery of spurious circular RNAs. The significance of this problem was highlighted in a recent survey of circular RNA in Drosophila, where non-statistical alignment procedures produced millions of candidate circular RNAs which those authors believed to be dominated by artifacts [8]. High confidence reconstruction of mRNA splicing events in linear RNAs from short read RNA-Seq data remains a challenging and unsolved problem, despite advances in this area. Detection of circular RNA faces the same challenges as well as additional concerns, including systematic biases in many widely used sequencing techniques and algorithms that can lead to characteristic false positive or false negative calls. For example, some approaches will be biased against detecting very small exonic circles because read lengths exceed the length of these circular RNAs. Such an ascertainment bias might negatively impact results from genome-wide approaches to test the hypothesis that circular RNAs containing a single exon tend to be longer than other circularized exons and are therefore more efficiently circularized, as was suggested in [9]. In this paper, we present a new statistical algorithm for quantifying the expression of circular and linear RNA spliced at annotated and un-annotated exon boundaries. This new algorithm assigns confidence (a posterior probability that a junction is expressed, and a p value for this probability) for each detected circular or linear RNA junction from RNA-Seq experiments. It differs from other published methods in that it calculates a statistical score for each read based on alignment properties, including the number of mismatches and mapping quality. These scores are aggregated for all reads that span a putative junction to assess the strength of evidence that this junction is expressed, a concept which we previously used in a much more simple form to detect circular RNAs only at annotated exonic boundaries [3]. We demonstrate a reduction of false positive and negative results compared with other methods, and show that our improved accuracy can have significant implications for genome-wide analysis. While we focus on the methodological approach applied to circular RNA in humans in this paper, our algorithm is equally powerful and applicable to the study of linear mRNA splicing and can be applied to any genome. We used this algorithm to rapidly test large numbers of data sets for regulation of circular RNA. Motivated by the observation that a highly expressed circular RNA in the mouse, Sry, is induced during embryonic development [10], we included data from developmental time courses to test the hypothesis that developmental induction of circular RNA could be a more general phenomenon. Our algorithm allowed us to discover striking induction of circular RNA during weeks 10–20 of human fetal development, including in the heart and lung. We found particularly high levels of circular RNA isoforms in the developing brain, including in genes essential to neurogenesis, such as lncRNA RMST [11], adding to the recent report that circular RNA is enriched in the aging fly brain [8]. We have shown that induction of circular RNA in the heart is recapitulated in in vitro directed differentiation of human pluripotent stem cells to the cardiomyocyte lineage and focus on the circular RNA from the gene NCX1, a calcium transporter that is essential for cardiac development [12, 13]. The developmental induction of the circular RNA from NCX1 is conserved at least between human and mouse, suggesting that developmental regulation of circular RNA may be evolutionarily conserved. In addition to these results, our algorithm unveiled circular RNAs that use splice sites not annotated in the current genome databases, including some spliced by the U12 (minor) spliceosome that likely use splice sites that are not used for splicing of the linear mRNA. RESULTS Statistical annotation-dependent algorithm Our analysis pipeline (Fig. 1a) uses

Read the complete text