- freely available
Int. J. Mol. Sci. 2013, 14(10), 20820-20832; doi:10.3390/ijms141020820
Abstract: Current technologies that are used for genome-wide microRNA (miRNA) prediction are mainly based on BLAST tool. They often produce a large number of false positives. Here, we describe an effective approach for identifying orthologous pre-miRNAs in several primates based on syntenic information. Some of them have been validated by small RNA high throughput sequencing data. This approach uses the synteny information and experimentally validated miRNAs of human, and incorporates currently available algorithms and tools to identify the pre-miRNAs in five other primates. First, we identified 929 potential pre-miRNAs in the marmoset in which miRNAs have not yet been reported. Then, we predicted the miRNAs in other primates, and we successfully re-identified most of the published miRNAs and found 721, 979, 650 and 639 new potential pre-miRNAs in chimpanzee, gorilla, orangutan and rhesus macaque, respectively. Furthermore, the miRNA transcriptome in the four primates have been re-analyzed and some novel predicted miRNAs have been supported by the small RNA sequencing data. Finally, we analyzed the potential functions of those validated miRNAs and explored the regulatory elements and transcription factors of some validated miRNA genes of interest. The results show that our approach can effectively identify novel miRNAs and some miRNAs that supported by small RNA sequencing data maybe play roles in the nervous system.
MicroRNA (miRNA) is a vital class of regulated RNA. miRNAs regulate gene expression by base pairing to target mRNAs causing translation repression or mRNA degradation. miRNAs were identified usually by molecular cloning and sequencing technologies. These methods tend to be highly biased towards abundantly and/or ubiquitously expressed miRNAs [1–3] and, in addition, certain miRNAs are likely to be missed during sequence cloning procedures .
A number of computational methods to identify miRNAs in an organism have been published. Most of these methods use the BLAST program  to generate local sequence alignments that are used to identify homologous pre-miRNAs [5–7]; a few of these methods use the BLAT tool  to search for candidate pre-miRNAs [9,10]. One of the disadvantages of using BLAST is that the region that is aligned corresponds to an optimal local alignment. miRNAs from the same family often diverge at the 5′ and 3′ ends of the genes in distantly-related organisms. In such cases, the optimal local BLAST alignments do not extend enough to cover the entire length of the pre-miRNA. In a number of studies, when the short mature miRNAs sequences were used as query sequences in BLAST searches, two putative miRNA hairpin structural regions were extracted by extending the aligned sequence upstream and downstream of the BLAST alignments [6,11]. Although the BLAST method has high sensitivity, it may produce a high number of false positives because no potential transcriptional information which could be predicted as pre-miRNAs is available.
The rapid development of sequencing technology has led to more and more genomic sequences becoming available and web sites such as the UCSC Genome Browse  have published global alignments of the genomic data of two or more organisms [13,14]. Because most miRNAs are conserved, especially in two closely-related organisms, it should be possible to obtain the syntenic homologous miRNAs of an organism using comparative genomics if the miRNAs of another closely-related organism are known. Whole genome alignments should make it possible to use syntenic homologous blocks to predict pre-miRNAs, and some of the problems of using the prediction methods based on BLAST will be avoided, making the identification of pre-miRNAs more reliable.
More recently, the newly developed deep sequencing technologies have been used to detect small RNAs [15–20]. Deep sequencing is a very effective method for the large-scale discovery of small non-coding RNAs. Especially, this approach could identify poorly expressed or species-specific miRNAs . However, in many cases, deep sequencing is used to uncover miRNAs that are expressed in limited conditions. Some miRNAs are expressed in response to a particular external environmental stimulation, some are expressed at a specific developmental or disease stage, and others are expressed only in some tissues. Fortunately, we could use comparative genomics method to predict their homologs in our organism of interest based on some poorly expressed or species-specific miRNAs that have been identified in other organisms. In contrast with human, far less miRNAs in other primates have been documented. To uncover novel miRNAs in some primates, we discovered novel miRNAs by an effective computational approach based on syntenic information. Meanwhile, we re-analyzed small RNA sequencing data in four primates and the results also supported some computational results. Moreover, we analyzed the potential functions of those validated miRNAs and we also explored their transcription factor and corresponding binding sites among the potential regulatory sequences of some validated miRNA genes of interest.
2. Results and Discussion
2.1. Computational Identification of Marmoset Pre-miRNAs
Pre-miRNAs are usually located in highly conserved syntenic regions in closely related species. It enables identification of miRNAs based on synteny information among closely related species. To test the effectiveness of our method, we first used it to identify candidate pre-miRNAs in the marmoset genome with the well-studied human miRNA data (Table 1). Marmosets are arboreal primates and the smallest monkey in the world. As yet, no studies on marmoset miRNAs have been published. Fortunately, we identified 1288 candidate miRNA loci and sequences in marmoset using synteny information between marmoset and human genome. From these candidates, 929 of them were determined as potential pre-miRNAs by MiPred (Southeast University, Nanjing, China)  (Table 1; the sequences and genome coordinates are listed in Supplementary Information 1  and Table S1). Further analysis revealed that only 387 pre-miRNAs could be classified into 225 known miRNA families according to the miRNA family information in the current miRBase (The University of Manchester, Manchester, UK)  (Table S2). The miR-515 family with 28 pre-miRNA members is the biggest miRNA family that was annotated by the Infernal tool in the marmoset genome (Table S2). In the human genome, the largest miRNA family in the current miRBase is the miR-548 family with 69 pre-miRNA members; the second largest is the miR-515 family with 42 members. The Infernal tool lacks sufficient power to classify miR-548 family; only 12 members of the miR-548 family have been identified in humans, and none of the 31 miR-548 precursors have been identified in marmoset, although they are homologs of human pre-miRNAs. The sequences of these members show low complexity and they are derived from MADE1 element [25,26].
2.2. Computational Identification of Pre-miRNAs in the Other Four Primates
We also used the computational approach to predict the pre-miRNAs in four other primates: chimpanzee, gorilla, orangutan and rhesus macaque, which share a close genetic relationship with human and marmoset (Table 1). The already known miRNAs of the four species are documented in miRBase 19. We identified 1324 candidate pre-miRNAs in the chimpanzee genome that are homologs of the human pre-miRNAs and 601 of the 655 chimpanzee pre-miRNAs documented in miRBase 19 were re-identified in our study. Thus, our method successfully re-identified 91.76% of the known pre-miRNAs and a further 721 new chimpanzee pre-miRNAs were identified (Table 1, Supplementary Information 1 and Table S1). In the orangutan genome, 590 of the 633 (93.21%) known pre-miRNAs documented in miRBase 19 were re-identified and 650 new orangutan pre-miRNAs were identified (Table 1, Supplementary Information 1 and Table S1). Moreover, in gorilla and rhesus macaque, 301 of 322 (93.48%) and 480 of 535 (89.72%) known pre-miRNAs were re-identified respectively. In total, 979 new gollira pre-miRNAs and 639 new pre-miRNAs in the rhesus macque were discovered in this study (Table 1, Supplementary Information 1 and Table S1).
Among the newly found pre-miRNAs, only 46 pre-miRNAs of the chimpanzee have been annotated into 40 known miRNA families (Table S2), 197 pre-miRNAs of gorilla have been classified into 127 known miRNA families (Table S2), and 50 pre-miRNAs have been annotated into 38 known miRNA families in the orangutan while 50 pre-miRNAs have been classified into 34 known miRNA families in rhesus macaque (Table S2) by using the Infernal tool . After analyzing those newly identified pre-miRNAs, we also found there are 172 conservative pre-miRNAs across human, chimpanzee, gorilla, orangutan, rhesus macaque and marmoset (Table S3).
2.3. Novel Predicted Pre-miRNAs Validated by High Throughput Small RNA Sequencing Data
Here, the small RNA deep sequencing data is derived from the brain samples in chimpanzee and rhesus macaque and the brain and liver samples in gorilla and orangutan [28,29]. These data have been re-analyzed to discover novel miRNAs to support our previous computational results. After processing with miRDeep2 (The Max Delbrück Center for Molecular Medicine, Berlin, Germany)  and filtering with a positive score and a significant p-value, we found 234 new pre-miRNAs (66 with expression evidence for star sequence) in the chimpanzee, 237 (53 with expression evidence for star sequence) in the gorilla, 393 (67 with expression evidence for star sequence) in the orangutan and 230 (62 with expression evidence for star sequence) in the rhesus macaque (Tables 1, S4, Figure 1).
We found 56, 25, 16 and 29 predicted pre-miRNAs have been validated by the small RNA sequencing data in chimpanzee, gorilla, orangutan and rhesus macaque, respectively (Figure 1, Tables 1, S4 and S5). Those were far less than our previous predicted results (Figure 1, Table S1). The main reason is that miRNAs expression response to environmental stimulus in different tissues at different development stages or different pathological states, and the miRNAs were detected only from several tissues could not represent all the miRNAs in the whole body. In this work, the novel miRNAs and pre-miRNAs were identified in the miRNA transcriptom data (Table S4) which is only originated from the tissues of brain or liver [28,29], while the pre-miRNAs (Table S1) were predicted by comparative genomics method may express in various tissues at different developmental stages due to the origin of their homologs in human are diversity. Furthermore, the available miRNA transcriptome data of the four primates is also very limited.
2.4. The Interesting Functional and Biological Importance of These Validated miRNAs
As the small RNA deep sequencing data is derived from the brain samples in chimpanzee and rhesus macaque and the brain and liver samples in gorilla and orangutan, some validated miRNAs (Figure 1, Tables S4 and S5) maybe play roles in the nervous system. To explore the biological processes that the validated mRNAs in these four primates are involved, we obtained their predicted targets using TargetScan (Whitehead Institute for Biomedical Research, Cambridge, MA, USA)  and PITA (Weizmann Institute, Rehovot, Israel) . Moreover, 311, 471, 420 and 69 common targets of these miRNAs were identified in chimpanzee, gorilla, orangutan and rhesus macaque, respectively. After doing functional enrichment analyses of these common targets in each organism, we found especially the targets of 13 validated miRNAs (Table S6) in orangutan showed significant over-representation of 12 Gene ontology (GO) terms involved in nervous system, such as “neuron apoptotic process”, “regulation of neuron apoptotic process”, “negative regulation of neuron apoptotic process” and so on (Table S7). Moreover, in chimpanzee, those targets significantly involved in two GO terms involved in nervous system and they are “ciliary neurotrophic factor-mediated signaling pathway” and “glial cell differentiation” (Table S7) and the corresponding 11 miRNAs are listed in Table S6. However, for those validated miRNAs and their targets of the other two primates, we have not seen obvious signals involved in nervous system (Table S7).
2.5. The Potential Transcription Factor Binding Sites of the Validated miRNA Genes Which Involve in Nervous System
Since the target genes of the miRNAs (Table S6) have been inferred to play roles in the brain, could we find some brain-specific regulatory information among those miRNAs? As there lacks enough tissue expression information in chimpanzee and orangutan, it is hard to find their tissue-specific regulatory sequences, but we could maybe find some common transcription factor binding sites (TFBSs) between them. Previous studies showed some intronic miRNAs have independent regulatory elements while other intronic miRNAs share the regulatory elements with their host genes . Moreover, nucleosomes can inhibit or facilitate transcription factor (TF) binding, and TFBSs tend to be located in low nucleosome occupancy or nucleosome-depleted regions [34,35]. Thus, it was feasible to predict the TFs and TFBSs based on the sequences and their nucleosome position information. To do this, we first classified the miRNAs into intronic miRNAs and intergenic miRNAs based on their genomic contexts and we found 12 intronic miRNAs and 12 intergenic miRNAs among the 24 miRNAs (Tables 2, S6). Then, we scanned their potential upstream regulatory sequences of those miRNAs combining with the nucleosome-DNA interaction model [36,37] and vertebrate position frequency matrices (PFMs) from JASPAR (University of Copenhagen, Copenhagen, Denmark) . Moreover, in the chimpanzee, only the common binding sites of Hltf have been found locating in the upstream sequences of two out of three intergenic miRNA genes and no motifs have been found in the upstream sequence of intergenic ptr-mir-2964a (Table S6). ETS1, FOXL1 and Hltf are the three common TFs which binding sites are locating in the upstream sequences of six out of seven miRNA host genes in the chimpanzee (Tables S6, S8). Moreover, the binding sites of 19 TFs are located in the sequences between the start sites of the intron and the start sites of five intronic miRNA genes respectively and the five miRNAs are ptr-mir-2115, ptr-mir-2681, ptr-mir-4511, ptr-mir-4760 and ptr-mir-4762 (Tables S6, S8). However, no TFBSs have been found in the potential regulatory sequences of the other two intronic miRNAs: ptr-mir-548t and ptr-mir-4691 (Tables S6, S8). We inferred that the two intronic miRNA genes may share regulatory elements with their host genes and the 19 TFs of the intronic miRNAs are also including the previous Hltf, ETS1 and FOXL1. While in the orangutan, the binding sites of GATA2 have been discovered in the upstream sequences of the five intergenic miRNA genes except ppy-mir-200b, ppy-mir-203b and ppy-mir-3065 (Table S6). Eight common TFs: ETS1, FOXC1, FOXL1, GATA2, HOXA5, MZF1_1-4, Nkx2-5 and SPIB could bind the upstream sequences of the two out of three intronic miRNAs and no TFBSs have been found in the potential regulatory sequences of ppy-mir-301a and ppy-mir-4419a (Tables S6, S8). Moreover, ETS1, FOXC1, GATA3, Nkx2-5 and SOX10 could bind the upstream sequences of the three out of four intronic miRNAs (Tables S6, S8). In all, although we have determined these potential TFs and TFBSs by computational methods, the results with available experimental evidence are always more convincing.
3. Experimental Section
3.1. Data sources
The sequences and genome coordinates of the miRNAs used in this study were obtained from miRBase 19 . The LiftOver conversion data for the whole genome alignment of human sequence (reference sequence) with marmoset, chimpanzee, gorilla, orangutan and rhesus macaque were downloaded from UCSC Genome Browser . The LiftOver utility  translates genomic coordinates between assemblies.
The deep sequencing data of the chimpanzee and rhesus macaque was downloaded from NCBI GEO database (GSE26545) (National Center for Biotechnology Information, Bethesda, MD, USA) , and the small RNA sequencing data of the gorilla and orangutan came from European Bioinformatics Institute (Hinxton, Cambridge, UK) with the study accession number ERP000973 .
3.2. A Comparative Genomics Method for the Identification of Orthologous Pre-miRNAs
Human was selected as the reference organism because, to date, the human miRNAs are the most studied and have the most experimental evidence, and humans are closely related to other primates such as chimpanzee, gorilla, orangutan, rhesus macaque and marmoset. Using a comparative analysis method, the pre-miRNAs of marmoset, chimpanzee, gorilla, orangutan and rhesus macaqu have been investigated. The comparative genomics method in this study is explained as follows (see flow chart Figure 2): first, the LiftOver program  was used to map the genome coordinates of the human pre-miRNAs to the genomes of other closely related species to obtain the coordinates of the candidate homologous miRNAs on the genomes of the interested organisms; next, if different members of one miRNA family mapped to overlapping loci, then only one miRNA was selected to represent this miRNA family in that organism; after retrieving the sequences using the coordinates of the candidate pre-miRNAs they were classified using the MiPred program  with the modified threshold value of gene lengths based on the sequence data of animals that are available in miRBase; and the newly identified homologous sequences of the human pre-miRNAs were named according to the miRBase naming criteria . Finally, The potential pre-miRNAs were annotated and classified into the known miRNA families by program rfam_scan.pl, which is a perl wrapper for searching nucleotide sequences against the Rfam database (v11.0) Wellcome Trust Sanger Institute, Hinxton, Cambridge, UK)  using the Infernal software (HHMI’s Janelia Farm Research Campus, Ashburn, VA, USA) . Moreover, our scripts for the identification of orthologous pre-miRNAs are available in Supplementary information 2.
To test the effectiveness of the method, it was first applied to identify candidate marmoset miRNAs. Marmoset miRNAs are not available in miRBase and had never been reported yet. Moreover, we used it to identify miRNAs in chimpanzee, gorilla, orangutan and rhesus macaque, while some of their miRNAs have been reported and documented. The overlap ratio between the genome coordinates of computationally identified pre-miRNAs with the documented pre-miRNAs was calculated using the following formula:
when the overlap ratio is no less than 0.5, the pre-miRNA which overlapped with known pre-miRNA will be taken as a re-identified pre-miRNA.
3.3. Deep Sequencing Data Analysis and New miRNAs Identification
The small RNA sequencing data from chimpanzee, gorilla, orangutan and rhesus macaque [28,29] has been re-analyzed to discover novel miRNAs and also been used to validate the predicted results by the above method. These deep sequencing data were processed with miRDeep2  by using the parameters described in the previous reference . Comparing to the previous study , we added human and mouse mature miRNAs as the known mature sequences when running miRDeep2. Moreover, another difference is that human miRNA and pre-miRNA data has been updated in miRBase 19.
3.4. Target Gene Ontology Analysis and Transcription Factors and Transcription Factor Binding Sites Exploration
Targets of those validated miRNAs were predicted using TargetScan 5.2  and PITA  respectively. The intersection of TargetScan and PITA targets was used to do gene ontology annotation analysis. Enrichment of GO terms in miRNA predicted targets was performed using the GOstats Bioconductor package .
The transcription factors and transcription factor binding sites of those miRNAs which targets involve in the nervous system have been predicted by using the nucleosome-DNA interaction model [36,37] and vertebrate position frequency matrixs (PFMs) from JASPAR . The potential regulatory sequence is extracted from the start site of the intronic of the host coding-gene in which one miRNA locates to the start site of the pre-miRNA if the miRNA is an intronic miRNA, while if a miRNA is an intergenic miRNA or a protein-coding gene, the sequence will be extracted from the upstream 2000 bp of the start site of the pre-miRNA or the protein-coding gene. Moreover, the upstream sequence has been separated into multiple sequences by uncertainty “Ns” which are unknown bases in the upstream sequence before prediction.
In this report, we first introduced a new approach to identify the pre-miRNAs in a target organism. It takes advantage of the synteny information between a reference genome and a target genome to find the orthologous miRNAs, especially to identify orthologs of highly divergent miRNAs. With this method, we identified 929 potential pre-miRNAs in marmoset in which miRNAs have not yet been reported and we successfully found 721, 979, 650 and 639 new potential pre-miRNAs in chimpanzee, gorilla, orangutan and rhesus macaque, respectively. These data show that our approach can effectively identify pre-miRNAs by using the synteny information. Then 56, 25, 16 and 29 predicted pre-miRNAs have been validated by the small RNA sequencing data from the brain or liver tissues in chimpanzee, gorilla, orangutan and rhesus macaque, respectively. Moreover, we found some of the validated miRNAs may play roles in nervous system and we also found some of them could be regulated by some common TFs.
|Primates||miRBase 19 (Known miRNA Genes)||Candidate miRNA Genes||Supported by Small RNA Sequencing Data||New miRNA Genes Identified from Deep Sequencing Data|
|Re-identified miRNA Genes||Novel miRNA Genes|
This work was supported by a grant from the National Natural Science Foundation of China (Project No. 31100914), the National Basic Research Program of China (No. 2012CB316501), the National Natural Science Foundation of China (Project No. 31240080) and the Doctoral Initiating Project of Hunan University of Science and Technology (E51178). We would like to thank the three anonymous reviewers for their helpful comments.
Conflicts of Interest
The authors declare no conflict of interest.
- Bartel, D.P. MicroRNAs: Genomics, biogenesis, mechanism, and function. Cell 2004, 116, 281–297. [Google Scholar]
- Berezikov, E.; Cuppen, E.; Plasterk, R.H. Approaches to microRNA discovery. Nat. Genet 2006, 38, S2–S7. [Google Scholar]
- Jones-Rhoades, M.W.; Bartel, D.P.; Bartel, B. MicroRNAS and their regulatory roles in plants. Annu. Rev. Plant Biol 2006, 57, 19–53. [Google Scholar]
- Zhang, Z.; Schwartz, S.; Wagner, L.; Miller, W. A greedy algorithm for aligning DNA sequences. J. Comput. Biol 2000, 7, 203–214. [Google Scholar]
- Artzi, S.; Kiezun, A.; Shomron, N. miRNAminer: A tool for homologous microRNA gene search. BMC Bioinforma 2008. [Google Scholar] [CrossRef]
- Zhou, M.; Wang, Q.; Sun, J.; Li, X.; Xu, L.; Yang, H.; Shi, H.; Ning, S.; Chen, L.; Li, Y.; et al. In silico detection and characteristics of novel microRNA genes in the Equus caballus genome using an integrated ab initio and comparative genomic approach. Genomics 2009, 94, 125–131. [Google Scholar]
- Brameier, M. Genome-wide comparative analysis of microRNAs in three non-human primates. BMC Res. Notes 2010. [Google Scholar] [CrossRef]
- Kent, W.J. BLAT—The BLAST-like alignment tool. Genome Res 2002, 12, 656–664. [Google Scholar]
- Baev, V.; Daskalova, E.; Minkov, I. Computational identification of novel microRNA homologs in the chimpanzee genome. Comput. Biol. Chem 2009, 33, 62–70. [Google Scholar]
- Weber, M.J. New human and mouse microRNA genes found by homology search. FEBS J 2005, 272, 59–73. [Google Scholar]
- Luo, Y.; Zhang, S. Computational prediction of amphioxus microRNA genes and their targets. Gene 2009, 428, 41–46. [Google Scholar]
- Meyer, L.R.; Zweig, A.S.; Hinrichs, A.S.; Karolchik, D.; Kuhn, R.M.; Wong, M.; Sloan, C.A.; Rosenbloom, K.R.; Roe, G.; Rhead, B.; et al. The UCSC Genome Browser database: Extensions and updates 2013. Nucleic Acids Res 2013, 41, D64–D69. [Google Scholar]
- Miller, W.; Rosenbloom, K.; Hardison, R.C.; Hou, M.; Taylor, J.; Raney, B.; Burhans, R.; King, D.C.; Baertsch, R.; Blankenberg, D.; et al. 28-way vertebrate alignment and conservation track in the UCSC Genome Browser. Genome Res 2007, 17, 1797–1808. [Google Scholar]
- Schwartz, S.; Kent, W.J.; Smit, A.; Zhang, Z.; Baertsch, R.; Hardison, R.C.; Haussler, D.; Miller, W. Human-mouse alignments with BLASTZ. Genome Res 2003, 13, 103–107. [Google Scholar]
- Morin, R.D.; O’Connor, M.D.; Griffith, M.; Kuchenbauer, F.; Delaney, A.; Prabhu, A.L.; Zhao, Y.; McDonald, H.; Zeng, T.; Hirst, M.; et al. Application of massively parallel sequencing to microRNA profiling and discovery in human embryonic stem cells. Genome Res 2008, 18, 610–621. [Google Scholar]
- Landgraf, P.; Rusu, M.; Sheridan, R.; Sewer, A.; Iovino, N.; Aravin, A.; Pfeffer, S.; Rice, A.; Kamphorst, A.O.; Landthaler, M.; et al. A mammalian microRNA expression atlas based on small RNA library sequencing. Cell 2007, 129, 1401–1414. [Google Scholar]
- Friedländer, M.R.; Adamidi, C.; Han, T.; Lebedeva, S.; Isenbarger, T.A.; Hirst, M.; Marra, M.; Nusbaum, C.; Lee, W.L.; Jenkin, J.C.; et al. High-resolution profiling and discovery of planarian small RNAs. Proc. Natl. Acad. Sci. USA 2009, 106, 11546–11551. [Google Scholar]
- Zhang, G.; Guo, G.; Hu, X.; Zhang, Y.; Li, Q.; Li, R.; Zhuang, R.; Lu, Z.; He, Z.; Fang, X.; et al. Deep RNA sequencing at single base-pair resolution reveals high complexity of the rice transcriptome. Genome Res 2010, 20, 646–654. [Google Scholar]
- Joyce, C.E.; Zhou, X.; Xia, J.; Ryan, C.; Thrash, B.; Menter, A.; Zhang, W.; Bowcock, A.M. Deep sequencing of small RNAs from human skin reveals major alterations in the psoriasis miRNAome. Hum. Mol. Genet 2012, 20, 4025–4040. [Google Scholar]
- Dhahbi, J.M.; Atamna, H.; Boffelli, D.; Magis, W.; Spindler, S.R.; Martin, D.I. Deep sequencing reveals novel microRNAs and regulation of microRNA expression during cell senescence. PLoS One 2011, 6, e20509. [Google Scholar]
- Sunkar, R.; Zhou, X.; Zheng, Y.; Zhang, W.; Zhu, J.K. Identification of novel and candidate miRNAs in rice by high throughput sequencing. BMC Plant Biol 2008. [Google Scholar] [CrossRef]
- Jiang, P.; Wu, H.; Wang, W.; Ma, W.; Sun, X.; Lu, Z. MiPred: Classification of real and pseudo microRNA precursors using random forest prediction model with combined features. Nucleic Acids Res 2007, 35, W339–W344. [Google Scholar]
- Hofacker, I.L.; Fontana, W.; Stadler, P.F.; Bonhoeffer, L.S.; Tacker, M.; Schuster, P. Fast folding and comparison of RNA secondary structures. Chem. Mon 1994, 125, 167–188. [Google Scholar]
- Griffiths-Jones, S.; Grocock, R.J.; van Dongen, S.; Bateman, A.; Enright, A.J. miRBase: MicroRNA sequences, targets and gene nomenclature. Nucleic Acids Res 2006, 34, D140–D144. [Google Scholar]
- Piriyapongsa, J.; Jordan, I.K. A family of human microRNA genes from miniature inverted-repeat transposable elements. PLoS One 2007, 2, e203. [Google Scholar]
- Yuan, Z.; Sun, X.; Liu, H.; Xie, J. MicroRNA genes derived from repetitive elements and expanded by segmental duplication events in mammalian genomes. PLoS One 2011, 6, e17666. [Google Scholar]
- Nawrocki, E.P.; Kolbe, D.L.; Eddy, S.R. Infernal 1.0: Inference of RNA alignments. Bioinformatics 2009, 25, 1335–1337. [Google Scholar]
- Dannemann, M.; Nickel, B.; Lizano, E.; Burbano, H.A.; Kelso, J. Annotation of primate miRNAs by high throughput sequencing of small RNA libraries. BMC Genomics 2012, 13. [Google Scholar] [CrossRef]
- Hu, H.Y.; Guo, S.; Xi, J.; Yan, Z.; Fu, N.; Zhang, X.; Menzel, C.; Liang, H.; Yang, H.; Zhao, M.; et al. MicroRNA expression and regulation in human, chimpanzee, and macaque brains. PLoS Genet 2011, 7, e1002327. [Google Scholar]
- Friedlander, M.R.; Mackowiak, S.D.; Li, N.; Chen, W.; Rajewsky, N. miRDeep2 accurately identifies known and hundreds of novel microRNA genes in seven animal clades. Nucleic Acids Res 2012, 40, 37–52. [Google Scholar]
- Grimson, A.; Farh, K.K.; Johnston, W.K.; Garrett-Engele, P.; Lim, L.P.; Bartel, D.P. MicroRNA targeting specificity in mammals: determinants beyond seed pairing. Mol. Cell 2007, 27, 91–105. [Google Scholar]
- Kertesz, M.; Iovino, N.; Unnerstall, U.; Gaul, U.; Segal, E. The role of site accessibility in microRNA target recognition. Nat. Genet 2007, 39, 1278–1284. [Google Scholar]
- Ozsolak, F.; Poling, L.L.; Wang, Z.; Liu, H.; Liu, X.S.; Roeder, R.G.; Zhang, X.; Song, J.S.; Fisher, D.E. Chromatin structure analyses identify miRNA promoters. Genes Dev 2008, 22, 3172–3183. [Google Scholar]
- Daenen, F.; van Roy, F.; de Bleser, P.J. Low nucleosome occupancy is encoded around functional human transcription factor binding sites. BMC Genomics 2008, 9. [Google Scholar] [CrossRef]
- Wang, J.; Zhuang, J.; Iyer, S.; Lin, X.; Whitfield, T.W.; Greven, M.C.; Pierce, B.G.; Dong, X.; Kundaje, A.; Cheng, Y.; et al. Sequence features and chromatin structure around the genomic regions bound by 119 human transcription factors. Genome Res 2012, 22, 1798–1812. [Google Scholar]
- Kaplan, N.; Moore, I.K.; Fondufe-Mittendorf, Y.; Gossett, A.J.; Tillo, D.; Field, Y.; LeProust, E.M.; Hughes, T.R.; Lieb, J.D.; Widom, J.; et al. The DNA-encoded nucleosome organization of a eukaryotic genome. Nature 2009, 458, 362–366. [Google Scholar]
- Segal, E.; Fondufe-Mittendorf, Y.; Chen, L.; Thastrom, A.; Field, Y.; Moore, I.K.; Wang, J.P.; Widom, J. A genomic code for nucleosome positioning. Nature 2006, 442, 772–778. [Google Scholar]
- Portales-Casamar, E.; Thongjuea, S.; Kwon, A.T.; Arenillas, D.; Zhao, X.; Valen, E.; Yusuf, D.; Lenhard, B.; Wasserman, W.W.; Sandelin, A. JASPAR 2010: The greatly expanded open-access database of transcription factor binding profiles. Nucleic Acids Res 2010, 38, D105–D110. [Google Scholar]
- Rhead, B.; Karolchik, D.; Kuhn, R.M.; Hinrichs, A.S.; Zweig, A.S.; Fujita, P.A.; Diekhans, M.; Smith, K.E.; Rosenbloom, K.R.; Raney, B.J.; et al. The UCSC Genome Browser database: Update 2010. Nucleic Acids Res 2010, 38, D613–D619. [Google Scholar]
- Gardner, P.P.; Daub, J.; Tate, J.; Moore, B.L.; Osuch, I.H.; Griffiths-Jones, S.; Finn, R.D.; Nawrocki, E.P.; Kolbe, D.L.; Eddy, S.R.; et al. Rfam: Wikipedia, clans and the “decimal” release. Nucleic Acids Res 2011, 39, D141–D145. [Google Scholar]
- Falcon, S.; Gentleman, R. Using GOstats to test gene lists for GO term association. Bioinformatics 2007, 23, 257–258. [Google Scholar]
© 2013 by the authors; licensee MDPI, Basel, Switzerland This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/).