Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Genome-Wide Discovery of Long Non-Coding RNAs in Rainbow Trout

  • Rafet Al-Tobasei,

    Affiliation Computational Science Program, Middle Tennessee State University, Murfreesboro, TN, 37132, United States of America

  • Bam Paneru,

    Affiliation Department of Biology and Molecular Biosciences Program, Middle Tennessee State University, Murfreesboro, TN, 37132, United States of America

  • Mohamed Salem

    Mohamed.salem@mtsu.edu

    Affiliations Computational Science Program, Middle Tennessee State University, Murfreesboro, TN, 37132, United States of America, Department of Biology and Molecular Biosciences Program, Middle Tennessee State University, Murfreesboro, TN, 37132, United States of America

Genome-Wide Discovery of Long Non-Coding RNAs in Rainbow Trout

  • Rafet Al-Tobasei, 
  • Bam Paneru, 
  • Mohamed Salem
PLOS
x

Abstract

The ENCODE project revealed that ~70% of the human genome is transcribed. While only 1–2% of the RNAs encode for proteins, the rest are non-coding RNAs. Long non-coding RNAs (lncRNAs) form a diverse class of non-coding RNAs that are longer than 200nt. Emerging evidence indicates that lncRNAs play critical roles in various cellular processes including regulation of gene expression. LncRNAs show low levels of gene expression and sequence conservation, which make their computational identification in genomes difficult. In this study, more than two billion Illumina sequence reads were mapped to the genome reference using the TopHat and Cufflinks software. Transcripts shorter than 200nt, with more than 83–100 amino acids ORF, or with significant homologies to the NCBI nr-protein database were removed. In addition, a computational pipeline was used to filter the remaining transcripts based on a protein-coding-score test. Depending on the filtering stringency conditions, between 31,195 and 54,503 lncRNAs were identified, with only 421 matching known lncRNAs in other species. A digital gene expression atlas revealed 2,935 tissue-specific and 3,269 ubiquitously-expressed lncRNAs. This study annotates the lncRNA rainbow trout genome and provides a valuable resource for functional genomics research in salmonids.

Introduction

Global gene expression data in different mammalian species have demonstrated that protein-coding sequences occupy less than 2% of the genome, and the vast majority of the genome is transcribed into non-coding RNAs [14]. These non-coding RNA molecules include small nuclear RNA (snRNA), small nucleolar RNA (snoRNA), microRNA (miRNA), small interfering RNA (siRNA), piwi RNA (piRNA), signal recognition particle (SRP) RNA and lncRNA. LncRNAs are defined as non-protein-coding RNAs greater than 200 nucleotides in length, distinguishing them from small non-coding RNAs [5, 6]. Based on their proximity to the protein-coding genes in a genome, lncRNAs are subdivided as genic (intronic or exonic with sense, antisense, and bidirectional orientation) or intergenic [7, 8]. Unlike small non-coding RNAs, lncRNA sequences are less conserved and are expressed at relatively low levels, and these characteristics make their computational identification and annotation difficult [9].

Like protein-coding genes, lncRNAs are often transcribed by RNA polymerase II and can be post transcriptionally modified by splicing, capping and polyadenylation [1013]. In contrast to protein-coding genes, a majority of lncRNA transcripts tend to have fewer exons [9] and a shorter transcript size (average of 800 nucleotides) [14]. LncRNAs usually exhibit highly cell- and tissue-specific expression patterns and sometimes they are uniquely localized to a specific cellular compartment [1518].

Even though a small number of lncRNAs have experimentally validated molecular functions, a substantial number of lncRNAs have been functionally annotated. LncRNAs are considered important gene regulators due to, at least, three important molecular roles; these RNAs serve as decoys, scaffolds or guides. Many lncRNAs serve as decoys that preclude access to DNA by regulatory proteins; this role affects transcription of protein-coding genes [19, 20]. Some lncRNAs regulate genes by acting as scaffolds to bring two or more proteins into a discrete complex [2124]. Other lncRNAs regulate different developmental and cellular processes by guiding a specific protein complexes to a specific promoter in response to certain molecular signals [2527]. LncRNA mediated guidance of chromatin modifying proteins affects expression of neighboring genes (cis) or distant genes (trans) and there is evidence that even cis acting lncRNAs have ability to act in trans [2830]. Beside transcriptional control, lncRNAs regulate many molecular processes including alternative splicing [31, 32], other post transcriptional processes [33], and mRNA transport [34].

Aquaculture of rainbow trout supplies a significant portion of aquatic food in the USA and worldwide. In addition to its importance as a food species, rainbow trout is one of the most widely used fish species as a model in biomedical research [3542]. In order to improve aquaculture production and efficiency and facilitate biomedical research of involving rainbow trout, a great deal of genetic information has been accumulated for this species that includes a recently published initial draft of the genome [4] and several assemblies of the transcriptome [4345]. However, a complete understanding of the trout’s genome biology is still lacking. Recent studies in mammalian and non-mammalian species have resolved some long-standing mysteries in biology by functionally characterizing lncRNAs as important regulators of protein-coding genes [24, 4650]. With growing interest in lncRNAs-mediated gene regulation, these RNAs have been characterized, genome-wide, in limited animal and plant species in recent years [15, 51]. And, our knowledge of lncRNAs in fish is still very limited [52]. Therefore, the objective of this study was to identify and characterize lncRNAs in rainbow trout genome and create a global gene expression atlas of lncRNAs in several vital tissues.

Materials and Methods

Ethics Statement

Institutional Animal Care and Use Committee of The United States Department of Agriculture, National Center for Cool and Cold Water Aquaculture (Leetown, WV) specifically reviewed and approved all husbandry practices used in this study (IACUC approval #056).

Data Source

To facilitate lncRNA discovery in rainbow trout, four high-throughput sequence datasets were used in this study. 1) About 1.16 billion Illumina sequence reads as we previously described [43]. Briefly, 13 tissues including brain, white muscle, red muscle, fat, gill, head kidney, kidney, intestine, skin, spleen, stomach, liver and testis were sequenced from a single male-doubled haploid rainbow trout. Sequencing libraries were constructed using poly-A selection technique and cDNA libraries were sequenced using Illumina’s paired-end protocol. Data were generated from a single doubled haploid individual to overcome the assembly bioinformatics challenges of the trout duplicated genome. 2) Similarly, about 0.75 billion Illumina single reads, used in annotating the rainbow trout genome and sequenced from a doubled haploid female rainbow trout, as previously described by Berthelot et al.[53]. Briefly, 13 vital tissues including (liver, brain, heart, skin, ovary, white and red muscle, anterior and posterior kidney, pituitary gland, stomach, gills) were sequenced. Sequencing libraries were constructed using poly-A selection technique and cDNA libraries were sequenced using Illumina’s 101 base-lengths single read protocol. 3) About 0.25 billion reads used in assembling the anadromous steelhead (Oncorhynchus mykiss) transcriptome by Fox et al. [45]. 4) About 89 million reads data set from redband trout (Oncorhynchus mykiss) by Narum et al. [54]. Data from Narum et al. were chosen because Ribo-Zero RNA-Seq libraries were sequenced to capture both the polyadenylated and the non- polyadenylated RNAs with information about transcript strand orientation.

Computational Prediction Pipeline

Sequencing reads were mapped to the genome reference [4] using the TopHat and Cufflinks software packages [55]. An in house Perl script was written to filter the transcripts shorter than 200 nt. Several stages of filtration were performed to remove protein-coding transcripts and small non-coding RNAs. First, transcripts were searched against NCBI nr protein database (updated on 10/01/2014). All the transcripts which had an open reading frame more than 100 amino acids were removed. Next, protein-coding calculator (CPC) was used to remove any remaining potential protein-coding transcripts (Index value <-0.5) [56]. To remove other classes of RNAs (tRNA, rRNA, snoRNA, miRNA, siRNA and other small non-coding RNAs) transcripts were searched against multiple RNA databases including genomic tRNA database, mirBase, LSU (large subunit ribosomal RNA) and SSU (Small subunit ribosomal RNA) databases [5760]. Any transcripts which showed sequence similarity with any of these classes of RNAs with cut-off E value of ≤ 0.0001 were removed. After these filtration steps, putative lncRNA transcripts were searched against several noncoding-RNA databases to explore sequence similarity of putative rainbow trout lncRNAs transcripts to previously characterized lncRNAs in other species [52, 6165]. All prediction steps were applied independently to the four transcriptome datasets. All putative lncRNAs from all four datasets were blasted against each other. LncRNA which were identified in at least 2 of the 4 datasets were chosen for further analysis. Data set from Narum et al., is the only one with information about strand orientation [54]. To ensure correct sense and antisense orientations of lncRNAs from the other three sources, their strand orientation was assigned by matching to counterparts from Narum and coworkers (based on sequence similarity match of more than 95% and same genomic location coordinates). A total of 54,503 non-redundant lncRNAs were identified in this dataset.

For the extra filtration steps, more stringently selected lncRNAs, any putative lncRNA containing ORF covering more 35% of its length or more than 83 amino acid were filtered out [66]. In addition, the cut-off value for the CPC [56] was decreased from -0.5 to -1.0. Further, if any lncRNA overlapped with more than 100 nt with another lncRNA from a different dataset, we filtered out the shortest lncRNA. Furthermore, any lncRNA that overlapped with a protein-coding gene in the sense orientation was removed. Lastly, any single-exon lncRNA that was adjacent to a protein-coding gene within 500nt was removed.

Identification of Tissue Expression

For lncRNA tissue distribution, sequencing reads from 13 tissues were independently mapped to all putative lncRNAs and gene expression level were measured in terms of RPKM. House-keeping and tissue-specific genes were determined as we previously described [43].

Gene Clustering

Sequencing reads from each tissue were mapped to combined reference consisting of the lncRNAs and mRNAs from the rainbow trout genome reference [4]. Expression of lncRNAs and protein-coding genes was determined in terms of RPKM. Expression value of each transcripts in each tissue was normalized using a scaling method in CLC genomics workbench with mean as the normalization value. Normalized expression values of transcripts in each of the 13 studied tissues were used to cluster protein-coding genes and lncRNAs using a clustering feature in Multi-experiment Viewer (MeV) program [67, 68]. The minimum correlation threshold to generate clusters was 0.97.

Identification of Genomic Location of lncRNAs Relative to Neighboring Protein-Coding Genes

LncRNAs were classified based on their intersection or relative location to protein-coding genes using in-house Perl scripts using the rainbow trout genome data (downloaded from http://www.genoscope.cns.fr/trout/data/).

Results and Discussion

Identification of Putative lncRNAs in Rainbow Trout

The main objective of this study was to identify a comprehensive list of putative lncRNA genes in the rainbow trout genome. To accomplish this, we sequenced poly-A selected cDNA libraries using total RNA isolated from 13 tissues. Recently, we used the same sequencing data to identify protein-coding transcripts in the trout genome [43]. In this study, sequence data for about 1.167 billion, paired-end reads (100 nt) were mapped against a reference rainbow trout genome using the Cufflink and TopHat software [55, 69], resulting in 231,505 putative transcripts. Several filtration steps were used to distinguish lncRNAs in the transcript list by removing the protein-coding transcripts, pseudogenes and other classes of non-coding RNAs including rRNA, miRNA, tRNA, snRNA, snoRNA (Fig 1). First, all transcripts shorter than 200 nt were removed, and then transcripts with an open reading frame (ORF) longer than 100 amino acids were filtered out. Next, remaining transcripts were BLASTx searched against the NCBI non-redundant protein database to eliminate transcripts with sequence similarity to known proteins at a cut off E-value of ≤ 0.0001. To further filter remaining protein-coding transcripts, we used the Coding Potential Calculator (CPC) software that assesses quality and completeness of query ORF to proteins in the NCBI database using six biologically meaningful sequence features [56]. These filtration steps left 44,350 transcripts from this data set that had very little or no evidence of protein-coding ability. Because most of the small non-coding RNAs like miRNA and tRNA are shorter than 200 nt, the first filtration step should be enough to remove most of the small non-coding RNAs. To confirm removal of any remaining small non-coding RNAs (tRNA, rRNA, snoRNA, miRNA, siRNA and other small non-coding RNAs), transcripts were searched against multiple RNA databases including genomic tRNA database, mirBase, and LSU (large subunit ribosomal RNA) and SSU (Small subunit ribosomal RNA) databases [5760]. After application of the above filtration steps, we found 44,124 putative lncRNAs from our sequence data set (Salem et al., [43]). These lncRNAs exhibited little or no evidence of coding potential or belonging to other non-coding classes of RNA.

thumbnail
Fig 1. Bioinformatics pipeline used in prediction of rainbow trout lncRNAs.

LncRNAs were predicted from four different transcriptomic datasets, then all putative lncRNAs from all data were blasted against each other. A total of 54,503 non-redundant lncRNAs identified in at least 2 of the 4 data sets were chosen for further analyses in order to increase the confidence of lncRNA prediction. Vertical arrows are pointing toward the subsequent prediction and filtration steps of the workflow. First horizontal arrow pointing toward the right is referring to the number of initial transcripts predicted from the four datasets. Middle six horizontal arrows indicate the number of transcripts filtered at each step and the final horizontal arrow points to the number of putative lncRNAs with significant hits to noncoding-RNA databases from each dataset.

http://dx.doi.org/10.1371/journal.pone.0148940.g001

Because some of the lncRNAs are thought to be due to expression noise [70], we conceptualized that prediction of lncRNAs from different reliable data sources would be an important step in removing false lncRNAs. To achieve this goal, the same lncRNAs prediction pipeline was applied to discover putative lncRNAs from three other rainbow trout transcriptomic datasets that are available on NCBI (Fig 1). Those three sources were sequence data used by Berthelot et al. [4] in annotating the rainbow trout genome, a data set used by Fox et al. [45] in assembling the anadromous steelhead (Oncorhynchus mykiss) transcriptome and a data set from redband trout (Oncorhynchus mykiss) that was reported by Narum et al. [54]. Data from Narum et al. were particularly useful because Ribo-Zero RNA-Seq protocols were used which allow sequencing both the polyadenylated and the non- polyadenylated RNAs. In addition, the strand orientation sequence information was preserved. From these three sequence data sources, a total of 0.75B reads, 89M reads, and 0.25B reads were used in the prediction pipeline that yielded 51,882; 1,191; and 36,474 putative lncRNAs in the three datasets, respectively. LncRNAs predicted in at least 2 of the 4 data sets were considered for the subsequent analyses. After removal of redundant transcripts, we had a total of 54,503 putative lncRNAs. Fig 1 illustrates the bioinformatics pipeline used in prediction of lncRNAs in all four datasets, and Table 1 and S1 table report the number of putative lncRNAs predicted in each dataset. FASTA and gtf annotation files are available at http://www.animalgenome.org/repository/pub/MTSU2015.1014/.

thumbnail
Table 1. Number of lncRNA predicted in at least 2 of the 4 datasets and final numbers after merging and removal of redundant sequences.

http://dx.doi.org/10.1371/journal.pone.0148940.t001

To look for evolutionarily conserved lncRNAs in rainbow trout, all putative lncRNA transcripts (54,503) were searched against several noncoding-RNA databases (E ≤ 0.0001) [52, 6165]. Of those 54,503 lncRNAs, only 421 had sequence homology to lncRNAs from other species (S1 table). This low evolutionary conservation of lncRNAs is in agreement with previous reports [9].

Characterization of lncRNAs

Studies on mouse, zebra fish and maize have suggested that lncRNAs are shorter than protein-coding genes, have relatively fewer exons, and are expressed at a lower level [51, 52, 71]. Consistent with previous reports, our study indicates that trout lncRNAs were shorter (0.821 kb) than protein-coding genes (1.636 kb) (Fig 2). In addition, the average number of exons in lncRNAs was 1.14 compared to 4.75 in protein-coding genes. Unlike the trout protein-coding genes, ~90% of the trout lncRNAs had one exon. Fig 2 and Table 2 show distribution and number of exons in lncRNAs compared to protein-coding genes. Data regarding exon numbers in lncRNAs from different species are inconsistent. Similar to our findings, some plant and animal studies reported one-exon bias for lncRNAs [51, 72]. Conversely, some human studies showed a remarkable two-exon prevalence in the majority of lncRNAs [9]. Several reasons may explain these discrepancies including tissue variation, developmental stages, sequencing techniques and biases due to variations in number and length of genes in different species.

thumbnail
Fig 2. Distribution of sequence length of LncRNAs compared to protein-coding transcripts in rainbow trout.

LncRNAs were shorter than protein-coding genes with (0.821 kb) and (1.636 kb) average length in each, respectively (Left). Distribution of number of exons in LncRNAs compared to that of protein-coding genes. Most LncRNA transcripts (~90%) have only one exon whereas majority of the protein-coding transcripts tend to have two or more exons (Right).

http://dx.doi.org/10.1371/journal.pone.0148940.g002

thumbnail
Table 2. Number of exons and average length of lncRNAs in different data sets.

http://dx.doi.org/10.1371/journal.pone.0148940.t002

LncRNAs are classified, based on their intersection with protein-coding genes, as genic and intergenic [9]. Some of the lncRNAs are located in transcriptionally-active regions and influence expression of neighboring genes [8, 73]. Therefore, the genomic position of lncRNAs relative to protein-coding genes can possibly provide important clues about lncRNA-mediated regulation of protein-coding genes [74]. Our data indicate that 7,847 (14.4%) of the lncRNAs intersected with protein-coding gene and thus are called genic (Fig 3). Of these lncRNAs 4,697 (8.6%), were intronic lncRNAs, existing in introns of protein-coding genes but do not intersect with any exons, and 3,091 (5.6%) exonic, sharing at least part of a protein-coding exon. Among those lncRNAs, 248 were sense and 1,488 were antisense; and 6,052 lncRNAs had an unknown orientation. In addition, there were 59 lncRNAs that completely overlapped with a protein-coding gene by containing this protein-coding gene within its intron. Fig 3 and S1 table show classification and number of lncRNAs based on their intersection with protein-coding genes. There were 46,656 (85.6%) intergenic lncRNAs in the trout genome that did not intersect but were within 15 kb of the nearest protein-coding gene. Those intergenic lncRNAs were further divided into 3,588 convergent (same sense) and 3,428 divergent (opposite sense). Consistent with our study, previous reports in humans indicate that the majority of lncRNA transcripts do not intersect with protein-coding genes [9].

thumbnail
Fig 3. Classification of lncRNAs based on their intersection with protein-coding genes and number of lncRNAs in each class.

Diagram on the top is a visual illustration of each class of lncRNAs relative to nearest protein-coding gene(s) based on genomic position and direction of transcripts. Bottom Fig in tabular format presents number of different classes of lncRNAs from each class. Numbers inside brackets following data source references indicate total number of that particular class of lncRNAs. Letters C, D, S, AS and U indicate number of convergent, divergent, sense, anti-sense and transcripts with unknown directionality, respectively.

http://dx.doi.org/10.1371/journal.pone.0148940.g003

Expression of lncRNA in Different Tissues

A comparison of lncRNA expression to protein-coding genes showed that transcript abundance of lncRNAs is lower than that of protein coding genes. Average RPKM (Reads Per Million per Kilo-base) of the most abundant 40,000 transcripts was 3.49 and 15.69 in LncRNAs and protein-coding genes, respectively (Fig 4). Similar trends, showing lower lncRNAs expression in all human tissues compared to mRNAs, were reported [9].

thumbnail
Fig 4. RPKM comparison of protein-coding genes and lncRNAs.

Transcript abundance of lncRNAs is lower than that of protein-coding genes. Average RPKM of the most abundant 40,000 genes is 15.69 and 3.49 for protein coding genes and LncRNAs, respectively (Left). Number of tissue-specific lncRNAs and protein-coding genes in various tissues. Expression of lncRNAs and protein-coding genes showed similar patterns among different tissues (Right).

http://dx.doi.org/10.1371/journal.pone.0148940.g004

Evidence is clear that lncRNAs exhibit strict cell/tissue specificity and play a significant role in development and differentiation of tissues in plants and animals [15, 51]. Nonetheless, their tissue specificity and potential role in tissue development are not well studied in fish. Lack of sequence conservation of lncRNAs across diverse species demands study of their expression in vital tissues as a method to identify lncRNAs with tissue-specific roles in rainbow trout. In this study, lncRNA expression was studied in 13 vital tissues of rainbow trout. Out of 54,503 putative lncRNAs, 3,269 (~5.9%) exhibited expression across all tissues with a minimum RPKM value of 1.0 (S2 table). On the other hand, 2,935 tissue-specific lncRNAs (5.4%) were identified from 13 tissues (S3 table). In this report, transcripts were described as ‘tissue specific’ if their expression in one tissue was 8-fold or higher compared to the maximum value for any of the other 12 tissues with a minimum RPKM of 0.5 [43] (Fig 4). Previously, we reported 17.1% and 8.9%, respectively, for housekeeping and tissue-specific protein-coding genes [43]. To gain insight into the expression and tissue specific differences between lncRNAs and protein-coding genes, the number of each was examined in 13 different tissues (Fig 4). Testis expressed the highest number of tissue-specific lncRNAs followed by brain, gill, and kidney. Conversely, liver expressed the lowest number of tissue-specific lncRNAs followed by skin, white muscle then spleen, in increasing order. We previously reported that the number of tissue-specific protein-coding transcripts follows similar patterns in various tissues [43]. Similar to the protein-coding genes, expression patterns of tissue-specific lncRNAs can be explained in terms of tissue complexity [43].

Previously, we showed that tissues are different in terms of the protein-coding transcriptome composition and complexity. Brain and testis possess the most complex transcriptomes. These tissues express large numbers of the genes; however, only a small part of the mRNA pool is expressed by the most abundant genes [43]. On the other hand, white muscle and stomach revealed simpler transcriptomes. These tissues express fewer genes and a greater proportion of the transcriptome comes from the most highly expressed genes. Similarly and in this study, complex tissues like brain and testis, expressed a larger number of lncRNAs with equal dominance of many transcripts (Fig 5). Conversely, white muscle, fat and liver showed less complex transcriptomes; a vast majority of the transcriptome included a few dominant lncRNAs. Similar expression patterns between protein-coding genes and lncRNAs may suggest common mechanisms of gene expression regulation and important role of lncRNAs in regulating protein-coding RNAs. Regardless, these data suggest that lncRNAs may be significant in determining tissue complexity.

thumbnail
Fig 5. Distribution of lncRNA expression in various tissues.

Proportion of the transcriptome that is contributed by the most abundant lncRNAs is plotted in various tissues. In complex tissues like brain and testis, larger number of lncRNAs were expressed with fairly equal dominance of many transcripts. On the contrary, less complex tissues like white muscle, fat and liver showed that majority of transcriptome is contributed by few dominant lncRNAs.

http://dx.doi.org/10.1371/journal.pone.0148940.g005

Correlation in Expression Patterns of lncRNA and Protein-Coding Genes across Tissues

Very low sequence conservation of lncRNAs hinders their molecular annotation. In order to look for possible functional significance of lncRNAs in regulating protein-coding genes, we constructed an expression-based relevance network between protein-coding genes and lncRNAs using a clustering algorithm in Multi-experiment Viewer software package (MeV) [67, 68]. In this study, biological correlation in expression patterns were compared across 13 tissues representing vastly different cellular and functional complexities. After clustering, genes of each cluster were ranked based on their entropies, and the top 20% of genes with the highest entropy were retained to construct networks. This approach identified 15 clusters containing protein-coding and lncRNA genes with strong correlation in their expression patterns (R2 >0.97) (S4 table). Examples of functionally important clusters include lncRNA Omy100084431 that was highly, positively correlated with splicing factor 3B (GSONMT00018324001) and transcription elongation factor SPT5 isoform X1 (GSONMT00067984001). In addition, expression of lncRNAs Omy200064145 and Omy100138726 was positively correlated with NF-kappa B inhibitor-like protein (GSONMT00082784001). Furthermore, a strong positive correlations in expression pattern between lncRNAs Omy300110093 and mitogen activated protein kinase1-like (GSONMT00053903001); Omy300072481 and thyroid hormone receptor alpha-like (GSONMT00066016001); Omy200106644 and histone deacetylase 3-like (GSONMT00058062001); and Omy300066671 and double-stranded RNA-specific adenosine deaminase (GSONMT00000999001) were observed. Proteins listed in these clusters have important functional roles in the cell including protein quality control (derlin-2) [75], RNA editing (adenosine deaminase) [76], transcriptional control (histone deacetylase 3) [77], splicing, and development. These findings nicely correlate with previously characterized molecular functions of lncRNAs in different species [23, 31, 32]. In order to explore additional underlying biological relationships between lncRNAs and protein-coding genes, more samples from different individuals and developmental stages should be studied as lncRNAs may be specific to developmental stages.

More Stringently Selected lncRNAs

The aforementioned 54,503 putative lncRNAs were identified using filtration steps with traditional cutoff values [52, 71]. To provide an optional more stringently selected list of lncRNAs, we performed extra filtration as follows. First, we calculated the average amino acid length for the shortest 10% of the rainbow trout protein-coding genes [42]; this calculation yielded 83 amino acids. Using 83 amino acids as the cut-off value of the lncRNA, 5,836 lncRNAs were filtered out of 54,503. In addition, lncRNA containing ORF covering more 35% of its length were filtered out [66]. Second, we decreased the cut-off value for the CPC [56] from -0.5 to -1.0, which filtered out an extra 4,978 leaving 43,689 putative lncRNA. The next filtration step was performed based on location of the lncRNAs in the genome predicted from a comparison of different datasets. If any lncRNA overlapped fully or partially by more than 100 nt with another lncRNA from a different dataset, we filtered out the shortest lncRNA; this step eliminated 5,945 putative lncRNAs. In addition, we filtered out any lncRNAs that overlapped with a protein-coding gene in the sense orientation and this filtration eliminated an additional 354 lncRNAs. The last filtration step removed any single-exonic lncRNA that was within 500 nt of a protein-coding gene; as a result, 1,538 putative lncRNAs were removed. The final number of putative lncRNAs was 31,195 (S1 table). FASTA and gtf annotation files are available at http://www.animalgenome.org/repository/pub/MTSU2015.1014/. Because the criteria for distinguishing lncRNAs are still loosely defined [78], filters applied in this study (with traditional or stringent cutoff values) should be considered arbitrary, hence, the identified lncRNAs may or may not reflect biological functions. For example, some of the well characterized lncRNAs in mammals contain more than 100 AA ORF. In this study, two sets of lncRNAs were obtained with traditional or stringent cut off values. All above mentioned analyses were done using lncRNAs from the traditional filtrations.

Supporting Information

S1 Table. Number, length, exon number, and genomic classification of putative lncRNAs predicted in four transcriptomic datasets.

doi:10.1371/journal.pone.0148940.s001

(XLSX)

S2 Table. Ubiquitously expressed lncRNAs.

doi:10.1371/journal.pone.0148940.s002

(XLSX)

S3 Table. Tissue-specific lncRNAs.

doi:10.1371/journal.pone.0148940.s003

(XLSX)

S4 Table. clusters containing protein-coding and lncRNA genes with strong correlation in their expression patterns.

doi:10.1371/journal.pone.0148940.s004

(XLSX)

Author Contributions

Conceived and designed the experiments: MS. Performed the experiments: RA BP. Analyzed the data: RA BP MS. Wrote the paper: BP MS RA.

References

  1. 1. Consortium EP. An integrated encyclopedia of DNA elements in the human genome. Nature. 2012;489(7414):57–74. doi: 10.1038/nature11247 pmid:22955616; PubMed Central PMCID: PMC3439153.
  2. 2. Clark MB, Choudhary A, Smith MA, Taft RJ, Mattick JS. The dark matter rises: the expanding world of regulatory RNAs. Essays Biochem. 2013;54:1–16. doi: 10.1042/bse0540001 pmid:23829523.
  3. 3. Djebali S, Davis CA, Merkel A, Dobin A, Lassmann T, Mortazavi A, et al. Landscape of transcription in human cells. Nature. 2012;489(7414):101–8. doi: 10.1038/nature11233 pmid:22955620; PubMed Central PMCID: PMCPMC3684276.
  4. 4. Berthelot C, Brunet F, Chalopin D, Juanchich A, Bernard M, Noel B, et al. The rainbow trout genome provides novel insights into evolution after whole-genome duplication in vertebrates. Nature communications. 2014;5:3657. doi: 10.1038/ncomms4657 pmid:24755649; PubMed Central PMCID: PMC4071752.
  5. 5. Zhu QH, Wang MB. Molecular Functions of Long Non-Coding RNAs in Plants. Genes (Basel). 2012;3(1):176–90. doi: 10.3390/genes3010176 pmid:24704849; PubMed Central PMCID: PMCPMC3899965.
  6. 6. Rinn JL, Chang HY. Genome regulation by long noncoding RNAs. Annu Rev Biochem. 2012;81:145–66. doi: 10.1146/annurev-biochem-051410-092902 pmid:22663078; PubMed Central PMCID: PMCPMC3858397.
  7. 7. Gibb EA, Brown CJ, Lam WL. The functional role of long non-coding RNA in human carcinomas. Mol Cancer. 2011;10:38. doi: 10.1186/1476-4598-10-38 pmid:21489289; PubMed Central PMCID: PMCPMC3098824.
  8. 8. Ponting CP, Oliver PL, Reik W. Evolution and functions of long noncoding RNAs. Cell. 2009;136(4):629–41. doi: 10.1016/j.cell.2009.02.006 pmid:19239885.
  9. 9. Derrien T, Johnson R, Bussotti G, Tanzer A, Djebali S, Tilgner H, et al. The GENCODE v7 catalog of human long noncoding RNAs: analysis of their gene structure, evolution, and expression. Genome research. 2012;22(9):1775–89. doi: 10.1101/gr.132159.111 pmid:22955988; PubMed Central PMCID: PMC3431493.
  10. 10. Guttman M, Amit I, Garber M, French C, Lin MF, Feldser D, et al. Chromatin signature reveals over a thousand highly conserved large non-coding RNAs in mammals. Nature. 2009;458(7235):223–7. doi: 10.1038/nature07672 pmid:19182780; PubMed Central PMCID: PMCPMC2754849.
  11. 11. Guttman M, Rinn JL. Modular regulatory principles of large non-coding RNAs. Nature. 2012;482(7385):339–46. doi: 10.1038/nature10887 pmid:22337053; PubMed Central PMCID: PMCPMC4197003.
  12. 12. Beaulieu YB, Kleinman CL, Landry-Voyer AM, Majewski J, Bachand F. Polyadenylation-dependent control of long noncoding RNA expression by the poly(A)-binding protein nuclear 1. PLoS Genet. 2012;8(11):e1003078. doi: 10.1371/journal.pgen.1003078 pmid:23166521; PubMed Central PMCID: PMCPMC3499365.
  13. 13. Yin QF, Yang L, Zhang Y, Xiang JF, Wu YW, Carmichael GG, et al. Long noncoding RNAs with snoRNA ends. Mol Cell. 2012;48(2):219–30. doi: 10.1016/j.molcel.2012.07.033 pmid:22959273.
  14. 14. Ørom UA, Derrien T, Beringer M, Gumireddy K, Gardini A, Bussotti G, et al. Long noncoding RNAs with enhancer-like function in human cells. Cell. 2010;143(1):46–58. doi: 10.1016/j.cell.2010.09.001 pmid:20887892; PubMed Central PMCID: PMCPMC4108080.
  15. 15. Cabili MN, Trapnell C, Goff L, Koziol M, Tazon-Vega B, Regev A, et al. Integrative annotation of human large intergenic noncoding RNAs reveals global properties and specific subclasses. Genes Dev. 2011;25(18):1915–27. doi: 10.1101/gad.17446611 pmid:21890647; PubMed Central PMCID: PMCPMC3185964.
  16. 16. Mercer TR, Dinger ME, Sunkin SM, Mehler MF, Mattick JS. Specific expression of long noncoding RNAs in the mouse brain. Proc Natl Acad Sci U S A. 2008;105(2):716–21. doi: 10.1073/pnas.0706729105 pmid:18184812; PubMed Central PMCID: PMCPMC2206602.
  17. 17. Prasanth KV, Prasanth SG, Xuan Z, Hearn S, Freier SM, Bennett CF, et al. Regulating gene expression through RNA nuclear retention. Cell. 2005;123(2):249–63. doi: 10.1016/j.cell.2005.08.033 pmid:16239143.
  18. 18. Ginger MR, Shore AN, Contreras A, Rijnkels M, Miller J, Gonzalez-Rimbau MF, et al. A noncoding RNA is a potential marker of cell fate during mammary gland development. Proc Natl Acad Sci U S A. 2006;103(15):5781–6. doi: 10.1073/pnas.0600745103 pmid:16574773; PubMed Central PMCID: PMCPMC1420634.
  19. 19. Hung T, Wang Y, Lin MF, Koegel AK, Kotake Y, Grant GD, et al. Extensive and coordinated transcription of noncoding RNAs within cell-cycle promoters. Nat Genet. 2011;43(7):621–9. doi: 10.1038/ng.848 pmid:21642992; PubMed Central PMCID: PMCPMC3652667.
  20. 20. Kino T, Hurt DE, Ichijo T, Nader N, Chrousos GP. Noncoding RNA gas5 is a growth arrest- and starvation-associated repressor of the glucocorticoid receptor. Sci Signal. 2010;3(107):ra8. doi: 10.1126/scisignal.2000568 pmid:20124551; PubMed Central PMCID: PMCPMC2819218.
  21. 21. Tsai MC, Manor O, Wan Y, Mosammaparast N, Wang JK, Lan F, et al. Long noncoding RNA as modular scaffold of histone modification complexes. Science. 2010;329(5992):689–93. doi: 10.1126/science.1192002 pmid:20616235; PubMed Central PMCID: PMCPMC2967777.
  22. 22. Pandey RR, Mondal T, Mohammad F, Enroth S, Redrup L, Komorowski J, et al. Kcnq1ot1 antisense noncoding RNA mediates lineage-specific transcriptional silencing through chromatin-level regulation. Mol Cell. 2008;32(2):232–46. doi: 10.1016/j.molcel.2008.08.022 pmid:18951091.
  23. 23. Yap KL, Li S, Muñoz-Cabello AM, Raguz S, Zeng L, Mujtaba S, et al. Molecular interplay of the noncoding RNA ANRIL and methylated histone H3 lysine 27 by polycomb CBX7 in transcriptional silencing of INK4a. Mol Cell. 2010;38(5):662–74. doi: 10.1016/j.molcel.2010.03.021 pmid:20541999; PubMed Central PMCID: PMCPMC2886305.
  24. 24. Kotake Y, Nakagawa T, Kitagawa K, Suzuki S, Liu N, Kitagawa M, et al. Long non-coding RNA ANRIL is required for the PRC2 recruitment to and silencing of p15(INK4B) tumor suppressor gene. Oncogene. 2011;30(16):1956–62. doi: 10.1038/onc.2010.568 pmid:21151178; PubMed Central PMCID: PMCPMC3230933.
  25. 25. Huarte M, Guttman M, Feldser D, Garber M, Koziol MJ, Kenzelmann-Broz D, et al. A large intergenic noncoding RNA induced by p53 mediates global gene repression in the p53 response. Cell. 2010;142(3):409–19. doi: 10.1016/j.cell.2010.06.040 pmid:20673990; PubMed Central PMCID: PMCPMC2956184.
  26. 26. Gupta RA, Shah N, Wang KC, Kim J, Horlings HM, Wong DJ, et al. Long non-coding RNA HOTAIR reprograms chromatin state to promote cancer metastasis. Nature. 2010;464(7291):1071–6. doi: 10.1038/nature08975 pmid:20393566; PubMed Central PMCID: PMCPMC3049919.
  27. 27. Rinn JL, Kertesz M, Wang JK, Squazzo SL, Xu X, Brugmann SA, et al. Functional demarcation of active and silent chromatin domains in human HOX loci by noncoding RNAs. Cell. 2007;129(7):1311–23. doi: 10.1016/j.cell.2007.05.022 pmid:17604720; PubMed Central PMCID: PMCPMC2084369.
  28. 28. Schmitz KM, Mayer C, Postepska A, Grummt I. Interaction of noncoding RNA with the rDNA promoter mediates recruitment of DNMT3b and silencing of rRNA genes. Genes Dev. 2010;24(20):2264–9. doi: 10.1101/gad.590910 pmid:20952535; PubMed Central PMCID: PMCPMC2956204.
  29. 29. Martianov I, Ramadass A, Serra Barros A, Chow N, Akoulitchev A. Repression of the human dihydrofolate reductase gene by a non-coding interfering transcript. Nature. 2007;445(7128):666–70. doi: 10.1038/nature05519 pmid:17237763.
  30. 30. Jeon Y, Lee JT. YY1 tethers Xist RNA to the inactive X nucleation center. Cell. 2011;146(1):119–33. doi: 10.1016/j.cell.2011.06.026 pmid:21729784; PubMed Central PMCID: PMCPMC3150513.
  31. 31. Tripathi V, Ellis JD, Shen Z, Song DY, Pan Q, Watt AT, et al. The nuclear-retained noncoding RNA MALAT1 regulates alternative splicing by modulating SR splicing factor phosphorylation. Mol Cell. 2010;39(6):925–38. doi: 10.1016/j.molcel.2010.08.011 pmid:20797886; PubMed Central PMCID: PMCPMC4158944.
  32. 32. Zong X, Tripathi V, Prasanth KV. RNA splicing control: yet another gene regulatory role for long nuclear noncoding RNAs. RNA Biol. 2011;8(6):968–77. doi: 10.4161/rna.8.6.17606 pmid:21941126; PubMed Central PMCID: PMCPMC3256421.
  33. 33. Yoon JH, Abdelmohsen K, Gorospe M. Posttranscriptional gene regulation by long noncoding RNA. J Mol Biol. 2013;425(19):3723–30. doi: 10.1016/j.jmb.2012.11.024 pmid:23178169; PubMed Central PMCID: PMCPMC3594629.
  34. 34. Tripathi V, Song DY, Zong X, Shevtsov SP, Hearn S, Fu XD, et al. SRSF1 regulates the assembly of pre-mRNA processing factors in nuclear speckles. Mol Biol Cell. 2012;23(18):3694–706. doi: 10.1091/mbc.E12-03-0206 pmid:22855529; PubMed Central PMCID: PMCPMC3442416.
  35. 35. Papanastasiou AD, Georgaka E, Zarkadis IK. Cloning of a CD59-like gene in rainbow trout. Expression and phylogenetic analysis of two isoforms. Molecular immunology. 2007;44(6):1300–6. doi: 10.1016/j.molimm.2006.05.014 pmid:16876248.
  36. 36. Williams DE. The rainbow trout liver cancer model: response to environmental chemicals and studies on promotion and chemoprevention. Comp Biochem Physiol C Toxicol Pharmacol. 2012;155(1):121–7. doi: 10.1016/j.cbpc.2011.05.013 pmid:21704190; PubMed Central PMCID: PMC3219792.
  37. 37. Giaquinto PC, Hara TJ. Discrimination of bile acids by the rainbow trout olfactory system: evidence as potential pheromone. Biological research. 2008;41(1):33–42. pmid:18769761.
  38. 38. McLAREN BA, O'DONNELL DJ, ELVEHJEM CA. Nutrition of rainbow trout. Fed Proc. 1947;6(1):413. pmid:20343876.
  39. 39. Patel M, Rogers JT, Pane EF, Wood CM. Renal responses to acute lead waterborne exposure in the freshwater rainbow trout (Oncorhynchus mykiss). Aquat Toxicol. 2006;80(4):362–71. doi: 10.1016/j.aquatox.2006.09.012 pmid:17125852.
  40. 40. Welsh PG, Lipton J, Mebane CA, Marr JC. Influence of flow-through and renewal exposures on the toxicity of copper to rainbow trout. Ecotoxicology and environmental safety. 2008;69(2):199–208. doi: 10.1016/j.ecoenv.2007.04.003 pmid:17517436.
  41. 41. Speare D, Arsenault G, Buote M. Evaluation of Rainbow Trout as a Model for use in Studies on Pathogenesis of the Branchial Microsporidian Loma salmonae. Contemporary topics in laboratory animal science / American Association for Laboratory Animal Science. 1998;37(2):55–8. pmid:12456170.
  42. 42. Davidson WS. Adaptation genomics: next generation sequencing reveals a shared haplotype for rapid early development in geographically and genetically distant populations of rainbow trout. Mol Ecol. 2012;21(2):219–22. pmid:22329016.
  43. 43. Salem M, Paneru B, Al-Tobasei R, Abdouni F, Thorgaard GH, Rexroad CE, et al. Transcriptome assembly, gene annotation and tissue gene expression atlas of the rainbow trout. PLoS One. 2015;10(3):e0121778. doi: 10.1371/journal.pone.0121778 pmid:25793877; PubMed Central PMCID: PMC4368115.
  44. 44. Salem M, Rexroad CE 3rd, Wang J, Thorgaard GH, Yao J. Characterization of the rainbow trout transcriptome using Sanger and 454-pyrosequencing approaches. BMC Genomics. 2010;11:564. Epub 2010/10/15. 1471-2164-11-564 [pii] doi: 10.1186/1471-2164-11-564 pmid:20942956; PubMed Central PMCID: PMC3091713.
  45. 45. Fox SE, Christie MR, Marine M, Priest HD, Mockler TC, Blouin MS. Sequencing and characterization of the anadromous steelhead (Oncorhynchus mykiss) transcriptome. Marine genomics. 2014;15:13–5. doi: 10.1016/j.margen.2013.12.001 pmid:24440488.
  46. 46. Kambara H, Niazi F, Kostadinova L, Moonka DK, Siegel CT, Post AB, et al. Negative regulation of the interferon response by an interferon-induced long non-coding RNA. Nucleic Acids Res. 2014;42(16):10668–80. doi: 10.1093/nar/gku713 pmid:25122750; PubMed Central PMCID: PMCPMC4176326.
  47. 47. Yang Z, Zhou L, Wu LM, Lai MC, Xie HY, Zhang F, et al. Overexpression of long non-coding RNA HOTAIR predicts tumor recurrence in hepatocellular carcinoma patients following liver transplantation. Ann Surg Oncol. 2011;18(5):1243–50. doi: 10.1245/s10434-011-1581-y pmid:21327457.
  48. 48. Kretz M, Siprashvili Z, Chu C, Webster DE, Zehnder A, Qu K, et al. Control of somatic tissue differentiation by the long non-coding RNA TINCR. Nature. 2013;493(7431):231–5. doi: 10.1038/nature11661 pmid:23201690; PubMed Central PMCID: PMCPMC3674581.
  49. 49. Luo M, Li Z, Wang W, Zeng Y, Liu Z, Qiu J. Long non-coding RNA H19 increases bladder cancer metastasis by associating with EZH2 and inhibiting E-cadherin expression. Cancer Lett. 2013;333(2):213–21. doi: 10.1016/j.canlet.2013.01.033 pmid:23354591.
  50. 50. Carrieri C, Cimatti L, Biagioli M, Beugnet A, Zucchelli S, Fedele S, et al. Long non-coding antisense RNA controls Uchl1 translation through an embedded SINEB2 repeat. Nature. 2012;491(7424):454–7. doi: 10.1038/nature11508 pmid:23064229.
  51. 51. Li L, Eichten SR, Shimizu R, Petsch K, Yeh CT, Wu W, et al. Genome-wide discovery and characterization of maize long non-coding RNAs. Genome Biol. 2014;15(2):R40. doi: 10.1186/gb-2014-15-2-r40 pmid:24576388; PubMed Central PMCID: PMC4053991.
  52. 52. Pauli A, Valen E, Lin MF, Garber M, Vastenhouw NL, Levin JZ, et al. Systematic identification of long noncoding RNAs expressed during zebrafish embryogenesis. Genome research. 2012;22(3):577–91. doi: 10.1101/gr.133009.111 pmid:22110045; PubMed Central PMCID: PMC3290793.
  53. 53. Berthelot C, Brunet F, Chalopin D, Juanchich A, Bernard M, Noël B, et al. The rainbow trout genome provides novel insights into evolution after whole-genome duplication in vertebrates. Nature communications. 2014;5:3657. doi: 10.1038/ncomms4657 pmid:24755649; PubMed Central PMCID: PMCPMC4071752.
  54. 54. Narum SR, Campbell NR. Transcriptomic response to heat stress among ecologically divergent populations of redband trout. BMC Genomics. 2015;16:103. doi: 10.1186/s12864-015-1246-5 pmid:25765850; PubMed Central PMCID: PMC4337095.
  55. 55. Trapnell C, Roberts A, Goff L, Pertea G, Kim D, Kelley DR, et al. Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks. Nature protocols. 2012;7(3):562–78. doi: 10.1038/nprot.2012.016 pmid:22383036; PubMed Central PMCID: PMC3334321.
  56. 56. Kong L, Zhang Y, Ye ZQ, Liu XQ, Zhao SQ, Wei L, et al. CPC: assess the protein-coding potential of transcripts using sequence features and support vector machine. Nucleic Acids Res. 2007;35(Web Server issue):W345–9. doi: 10.1093/nar/gkm391 pmid:17631615; PubMed Central PMCID: PMCPMC1933232.
  57. 57. Chan PP, Lowe TM. GtRNAdb: a database of transfer RNA genes detected in genomic sequence. Nucleic Acids Res. 2009;37(Database issue):D93–7. doi: 10.1093/nar/gkn787 pmid:18984615; PubMed Central PMCID: PMC2686519.
  58. 58. Wuyts J, Van de Peer Y, Winkelmans T, De Wachter R. The European database on small subunit ribosomal RNA. Nucleic Acids Res. 2002;30(1):183–5. pmid:11752288; PubMed Central PMCID: PMC99113.
  59. 59. Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 2013;41(Database issue):D590–6. doi: 10.1093/nar/gks1219 pmid:23193283; PubMed Central PMCID: PMC3531112.
  60. 60. Van Peer G, Lefever S, Anckaert J, Beckers A, Rihani A, Van Goethem A, et al. miRBase Tracker: keeping track of microRNA annotation changes. Database (Oxford). 2014;2014. doi: 10.1093/database/bau080 pmid:25157074; PubMed Central PMCID: PMC4142392.
  61. 61. Bu D, Yu K, Sun S, Xie C, Skogerbo G, Miao R, et al. NONCODE v3.0: integrative annotation of long noncoding RNAs. Nucleic Acids Res. 2012;40(Database issue):D210–5. doi: 10.1093/nar/gkr1175 pmid:22135294; PubMed Central PMCID: PMC3245065.
  62. 62. Kaushik K, Leonard VE, Kv S, Lalwani MK, Jalali S, Patowary A, et al. Dynamic expression of long non-coding RNAs (lncRNAs) in adult zebrafish. PLoS One. 2013;8(12):e83616. doi: 10.1371/journal.pone.0083616 pmid:24391796; PubMed Central PMCID: PMCPMC3877055.
  63. 63. Xie C, Yuan J, Li H, Li M, Zhao G, Bu D, et al. NONCODEv4: exploring the world of long non-coding RNA genes. Nucleic Acids Res. 2014;42(Database issue):D98–103. doi: 10.1093/nar/gkt1222 pmid:24285305; PubMed Central PMCID: PMCPMC3965073.
  64. 64. RNAcentral Consortium. RNAcentral: an international database of ncRNA sequences. Nucleic Acids Res. 2015;43(Database issue):D123–9. doi: 10.1093/nar/gku991 pmid:25352543; PubMed Central PMCID: PMCPMC4384043.
  65. 65. Quek XC, Thomson DW, Maag JL, Bartonicek N, Signal B, Clark MB, et al. lncRNAdb v2.0: expanding the reference database for functional long noncoding RNAs. Nucleic Acids Res. 2015;43(Database issue):D168–73. doi: 10.1093/nar/gku988 pmid:25332394; PubMed Central PMCID: PMCPMC4384040.
  66. 66. Ensembl. Annotation of Non-Coding RNAs [cited 2015 10/16]. Available: http://useast.ensembl.org/info/genome/genebuild/ncrna.html?redirect=no. Accessed 16 October 2015.
  67. 67. Howe EA, Sinha R, Schlauch D, Quackenbush J. RNA-Seq analysis in MeV. Bioinformatics. 2011;27(22):3209–10. doi: 10.1093/bioinformatics/btr490 pmid:21976420; PubMed Central PMCID: PMC3208390.
  68. 68. Saeed AI, Sharov V, White J, Li J, Liang W, Bhagabati N, et al. TM4: a free, open-source system for microarray data management and analysis. BioTechniques. 2003;34(2):374–8. pmid:12613259.
  69. 69. Langmead B, Salzberg SL. Fast gapped-read alignment with Bowtie 2. Nature methods. 2012;9(4):357–9. doi: 10.1038/nmeth.1923 pmid:22388286; PubMed Central PMCID: PMC3322381.
  70. 70. Louro R, Smirnova AS, Verjovski-Almeida S. Long intronic noncoding RNA transcription: expression noise or expression choice? Genomics. 2009;93(4):291–8. doi: 10.1016/j.ygeno.2008.11.009 pmid:19071207.
  71. 71. Zhang K, Huang K, Luo Y, Li S. Identification and functional analysis of long non-coding RNAs in mouse cleavage stage embryonic development based on single cell transcriptome data. BMC Genomics. 2014;15:845. doi: 10.1186/1471-2164-15-845 pmid:25277336; PubMed Central PMCID: PMC4200203.
  72. 72. Ravasi T, Suzuki H, Pang KC, Katayama S, Furuno M, Okunishi R, et al. Experimental validation of the regulated expression of large numbers of non-coding RNAs from the mouse genome. Genome research. 2006;16(1):11–9. doi: 10.1101/gr.4200206 pmid:16344565; PubMed Central PMCID: PMCPMC1356124.
  73. 73. Mercer TR, Dinger ME, Mattick JS. Long non-coding RNAs: insights into functions. Nat Rev Genet. 2009;10(3):155–9. doi: 10.1038/nrg2521 pmid:19188922.
  74. 74. Villegas VE, Zaphiropoulos PG. Neighboring gene regulation by antisense long non-coding RNAs. Int J Mol Sci. 2015;16(2):3251–66. doi: 10.3390/ijms16023251 pmid:25654223; PubMed Central PMCID: PMC4346893.
  75. 75. Dougan SK, Hu CC, Paquet ME, Greenblatt MB, Kim J, Lilley BN, et al. Derlin-2-deficient mice reveal an essential role for protein dislocation in chondrocytes. Mol Cell Biol. 2011;31(6):1145–59. doi: 10.1128/MCB.00967-10 pmid:21220515; PubMed Central PMCID: PMCPMC3067910.
  76. 76. Bass BL. RNA editing by adenosine deaminases that act on RNA. Annu Rev Biochem. 2002;71:817–46. doi: 10.1146/annurev.biochem.71.110601.135501 pmid:12045112; PubMed Central PMCID: PMCPMC1823043.
  77. 77. Wen YD, Perissi V, Staszewski LM, Yang WM, Krones A, Glass CK, et al. The histone deacetylase-3 complex contains nuclear receptor corepressors. Proc Natl Acad Sci U S A. 2000;97(13):7202–7. pmid:10860984; PubMed Central PMCID: PMCPMC16523.
  78. 78. Ulitsky I, Bartel DP. lincRNAs: genomics, evolution, and mechanisms. Cell. 2013;154(1):26–46. doi: 10.1016/j.cell.2013.06.020 pmid:23827673; PubMed Central PMCID: PMCPMC3924787.