Differentiating protein-coding and noncoding RNA: challenges and ambiguities.

Dinger ME, Pang KC, Mercer TR, Mattick JS. · 2008

Read Differentiating protein-coding and noncoding RNA: challenges and ambiguities. by Dinger ME, Pang KC, Mercer TR, Mattick JS. (2008) 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-pmc2518207.txt.

Introduction

Dinger, Pang, Mercer, and Mattick examined in 2008 a basic sorting problem in genetics: telling protein-coding RNA from noncoding RNA. For decades the split seemed easy, since messenger RNAs carry protein recipes and a few other RNAs had known jobs. Genome-wide studies then revealed thousands of transcripts that fit neither box cleanly.

The authors weigh the strategies used to classify transcripts and show where each one fails. Some RNAs labelled noncoding may actually encode small peptides, while some supposed messengers may never be translated. A few transcripts do both jobs at once, working as RNA and as a protein template. The review concludes that the two categories form a false split, and no transcript's RNA-level role should be dismissed.

Catalog record

CREATORDinger ME, Pang KC, Mercer TR, Mattick JS.
YEAR2008
COLLECTIONPAPERS
FORMATTXT
FILE SIZE35 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC2518207

Browse the full library · About the Tower of Babel

Full text

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

Differentiating Protein-Coding and Noncoding RNA: Challenges and Ambiguities By Marcel E Dinger, Ken C Pang, Tim R Mercer, John S Mattick, Johanna McEntyre Published in: PLoS Computational Biology DOI: 10.1371/journal.pcbi.1000176 ABSTRACT Abstract The assumption that RNA can be readily classified into either protein-coding or non-protein–coding categories has pervaded biology for close to 50 years. Until recently, discrimination between these two categories was relatively straightforward: most transcripts were clearly identifiable as protein-coding messenger RNAs (mRNAs), and readily distinguished from the small number of well-characterized non-protein–coding RNAs (ncRNAs), such as transfer, ribosomal, and spliceosomal RNAs. Recent genome-wide studies have revealed the existence of thousands of noncoding transcripts, whose function and significance are unclear. The discovery of this hidden transcriptome and the implicit challenge it presents to our understanding of the expression and regulation of genetic information has made the need to distinguish between mRNAs and ncRNAs both more pressing and more complicated. In this Review, we consider the diverse strategies employed to discriminate between protein-coding and noncoding transcripts and the fundamental difficulties that are inherent in what may superficially appear to be a simple problem. Misannotations can also run in both directions: some ncRNAs may actually encode peptides, and some of those currently thought to do so may not. Moreover, recent studies have shown that some RNAs can function both as mRNAs and intrinsically as functional ncRNAs, which may be a relatively widespread phenomenon. We conclude that it is difficult to annotate an RNA unequivocally as protein-coding or noncoding, with overlapping protein-coding and noncoding transcripts further confounding this distinction. In addition, the finding that some transcripts can function both intrinsically at the RNA level and to encode proteins suggests a false dichotomy between mRNAs and ncRNAs. Therefore, the functionality of any transcript at the RNA level should not be discounted. INTRODUCTION Numerous studies have demonstrated that the true catalog of RNAs encoded within the genome (the “transcriptome”) is more extensive and complex than previously thought (reviewed in [1]–[3]). In humans and mice, for instance, it has become apparent that the vast majority of the genome is transcribed, often in intricate networks of overlapping sense and antisense transcripts, many of which are alternatively spliced [1], [4]–[8]. However, mRNAs account for only ∼2.3% of the human genome [1],[9], and therefore the vast majority of this unexpected transcription, sometimes referred to as “dark matter” [10],[11], appears to be non-protein–coding. Unsurprisingly, a great deal of attention is now focused on the noncoding transcriptome. Dominating this field of inquiry has been the discovery of thousands of small RNAs (<200 nt in length). Many of these have since been classified into novel categories (e.g., microRNAs, PIWI-associated RNAs, and endogenous small interfering RNAs) on the basis of function, length, biogenesis, structural/sequence features, and protein-binding partners (reviewed in [12]). Interestingly, however, long ncRNAs (>200 nt) appear to comprise the largest portion of the mammalian noncoding transcriptome. Tiling array studies of the human genome, for instance, revealed that the majority of transcription occurs as long ncRNAs [13], some of which may be precursors for smaller RNAs, but many of which are detected as relatively stable polyadenylated and non-polyadenylated transcripts [8],[14]. The biological significance of these long ncRNAs is controversial. Despite an increasing number of long ncRNAs having been shown to fulfill a diverse range of regulatory roles (reviewed in [15],[16]), the functions of the vast majority remain unknown and untested. And while this is also true of small RNAs to some extent, long ncRNAs—unlike their smaller counterparts—lack obvious features to allow a priori functional categorization or prediction. Furthermore, the exact prevalence of long ncRNAs remains subject to significant interpretation and debate. For instance, the FANTOM and H-Invitational consortiums annotated comprehensive full-length cDNA collections in mouse and human, respectively [5],[17]. Despite using similar methods of cDNA library construction, the two groups came up with very different prevalence estimates for long ncRNAs within the mammalian genome: in mouse, 33% of transcripts (34,030/102,281) were annotated as noncoding; by comparison, only 7% of human transcripts (1,377/21,037) were identified as ncRNAs. That such divergent estimates exist highlights how difficult it has become to discriminate between long ncRNAs and mRNAs. To better understand the nature of these challenges, the approaches used to distinguish noncoding from protein-coding are considered below. STRATEGIES TO DISCRIMINATE BETWEEN NCRNAS AND MRNAS Open reading frame length One of the most fundamental criteria used to distinguish long ncRNAs from mRNAs is ORF length. Since short putative ORFs can be expected to occur by chance within long noncoding sequences, minimum ORF cutoffs are usually applied to reduce the likelihood of falsely categorizing ncRNAs as mRNAs. For instance, the FANTOM consortium originally used a cutoff of 300 nt (100 codons) to help identify putative mRNAs [18]. This somewhat arbitrary threshold is consistent with the observation that >95% of proteins in public databases such as Swiss-Prot and the International Protein Index are >100 aa in length [19], and has subsequently been shown to display a high level of concordance with more sophisticated discrimination methods [20]. This length is also approximately two standard deviations above the average length of ORFs in a one kilobase random sequence (Figure 1). [Figure: Incidence of open reading frames (ORFs) in randomly generated transcripts of increasing length.Twenty thousand transcripts of varying length and random nucleotide composition were computationally generated and scanned for ORFs. The maximum ORF and transcript lengths were plotted and fitted to a logarithmic curve. The shaded regions represent incidences of randomly occurring ORFs at 1, 2, or 3 standard deviations from the mean. The red line indicates the 300 nt ORF threshold that is commonly used to distinguish protein-coding genes in transcript classification pipelines. Therefore, this plot illustrates that for transcripts longer than ∼1000 bp, such a threshold may define transcripts as protein-coding that would be expected to occur by chance. The function y = 91.Ln(x)−330, which approximates random ORF incidence according to transcript length at two standard deviations above the mean (i.e., 95% confidence interval, indicated in green), could be used to discriminate noncoding from protein-coding transcripts in a transcript-length–dependent manner.] Using putative ORF length alone, although straightforward to apply across large datasets, is problematic for various reasons. First, bona fide long ncRNAs will by chance contain putative ORFs that are quite long. For instance, H19, Xist, Mirg, Gtl2, and KcnqOT1 all have putative ORFs >100 codons, but have been characterized as functional ncRNAs [15]. Applying a traditional ORF cutoff of 300 nt will therefore misclassify many ncRNAs as mRNAs, and this is especially true for very long ncRNAs, as illustrated in Figure 1. For example, murine Xist is ∼15 Kb in size [21] and contains a putative ORF of 298 aa, which led to the erroneous conclusion that it was a protein-coding gene when first discovered [22]. Second, with a cutoff of 300 nt, proteins <100 aa in size may also be incorrectly classified as ncRNAs. The potential scale of such errors is significant, given recent estimates that the mammalian proteome contains ∼3,700 proteins below this size [19]. To minimize such errors, ORF length cutoffs can be reduced, which is exactly what the H-Invitational consortium did in applying a threshold of 60 nt (20 codons) in their annotation pipeline [17]. However, such a low threshold will falsely underestimate

Read the complete text