New syntax to describe local continuous structure-sequence information for recognizing new pre-miRNAs

Wang, Minghao; Song, Xiaofeng; Han, Ping; Li, Wei; Jiang, Bin

doi:10.1016/j.jtbi.2010.02.037

Cited by 6 publications

(5 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, several studies have developed different ab initio miRNA prediction methods that are based mainly on characteristic features of miRNA, arising from the sequence and secondary structures. Examples of these are tripletSVM, 13 miPred 14 microPred, 15 MiPred, 16 ProMir, 17 SSCprofiler, 18,19 and the algorithm developed by Wang et al 20 Issues in the existing ab initio miRNA prediction methods include a poorer performance, an inability to being applied to multi-species data, and an inability to balance between specificity and sensitivity; these issues have not been addressed completely or successfully. For instance, tripletSVM, 13 which is a Support Vector Machine (SVM)-based classifier, achieves an average accuracy on a cross-species test of 90% but cannot predict multiloop hairpins.…”

Section: Resultsmentioning

confidence: 99%

“…The classifier MiPred 16 provides a web server to predict miRNAs but also does not support an analysis of pre-miRNA sequences with multiple loops. Wang et al 20 uses a new syntax to represent structure and sequence information, but it only achieves a sensitivity of 81.98% and a specificity of 87.16% on a human test set. More importantly, no existing methods provide an easy-to-use GUI program to predict miRNAs.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

mirExplorer: Detecting microRNAs from genome and next generation sequencing data using the AdaBoost method with transition probability matrix and combined features

Guan¹,

Liao²,

Qu³

et al. 2011

RNA Biology

View full text Add to dashboard Cite

microRNAs (miRNAs) represent an abundant group of small regulatory non-coding RNAs in eukaryotes. The emergence of Next-generation sequencing (NGS) technologies has allowed the systematic detection of small RNAs (sRNAs) and de novo sequencing of genomes quickly and with low cost. As a result, there is an increased need to develop fast miRNA prediction tools to annotate miRNAs from various organisms with a high level of accuracy, using the genome sequence or the NGS data. Several miRNA predictors have been proposed to achieve this purpose. However, the accuracy and fitness for multiple species of existing predictors needed to be improved. Here, we present a novel prediction tool called mirExplorer, which is based on an integrated adaptive boosting method and contains two modules. The first module named mirExplorer-genome was designed to de novo predict pre-miRNAs from genome, and the second module named mirExplorer-NGS was used to discover miRNAs from NGS data. A set of novel features of pre-miRNA secondary structure and miRNA biogenesis has been extracted to distinguish real pre-miRNAs from pseudo ones. We used outer-ten-fold cross-validation to verify the mirExplorer-genome computation, which obtained a specificity of 95.03% and a sensitivity of 93.71% on human data. This computation was made on test data from 16 species, and it achieved an overall accuracy of 95.53%. Systematic outer-ten-fold cross-validation of the mirExplorer-NGS model achieved a specificity of 98.3% and a sensitivity of 97.72%. We found that the good performance of the mirExplorer-NGS model was upheld across species from vertebrates to plants in test datasets. The mirExplorer is available as both web server and software package at http://biocenter.sysu.edu.cn/mir/.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

mirExplorer: Detecting microRNAs from genome and next generation sequencing data using the AdaBoost method with transition probability matrix and combined features

Guan¹,

Liao²,

Qu³

et al. 2011

RNA Biology

View full text Add to dashboard Cite

show abstract

“…The next generation of computational tools relied on more sophisticated machine learning algorithms such as support vector machines (SVMs) capable of taking into account multiple biological features such as free energy of the hairpin structure, paired bases, loop length, and stem conservation to predict novel miRNAs [19][20][21][22][23][24][25][26][27]. Two very effective computational studies utilized Hidden Markov Models (HMM) and a Bayesian classifier [28,29], to simultaneously consider sequence and structure features at the nucleotide level for predicting miRNA genes.…”

Section: Introductionmentioning

confidence: 99%

Finding Cancer-Associated miRNAs: Methods and Tools

et al. 2011

View full text Add to dashboard Cite

Changes in the structure and/or the expression of protein coding genes were thought to be the major cause of cancer for many decades. The recent discovery of non-coding RNA (ncRNA) transcripts (i.e., microRNAs) suggests that the molecular biology of cancer is far more complex. MicroRNAs (miRNAs) have been under investigation due to their involvement in carcinogenesis, often taking up roles of tumor suppressors or oncogenes. Due to the slow nature of experimental identification of miRNA genes, computational procedures have been applied as a valuable complement to cloning. Numerous computational tools, implemented to recognize the features of miRNA biogenesis, have resulted in the prediction of novel miRNA genes. Computational approaches provide clues as to which are the dominant features that characterize these regulatory units and furthermore act by narrowing down the search space making experimental verification faster and cheaper. In combination with large scale, high throughput methods, such as deep sequencing, computational methods have aided in the discovery of putative molecular signatures of miRNA deregulation in human tumors. This review focuses on existing computational methods for identifying miRNA genes, provides an overview of the methodology undertaken by these tools, and underlies their contribution towards unraveling the role of miRNAs in cancer.

show abstract

“…Within the RISC, miRNAs regulate gene expression by mainly in either way: mRNA cleavage or translational repression [1][2][3][4] . So far, thousands of miRNAs have been identified in animals, plants, and fungi by experiments or computational methods [5][6] . At the same time, each miRNA is thought to have hundreds of target mRNAs based on experimental or bioinformatics analysis [7][8][9] .…”

Section: Introductionmentioning

confidence: 99%

Computational analysis on the level of miRNA-mediated target mRNA degradation

Cheng

Zhang

Song

et al. 2011

Wuhan Univ. J. Nat. Sci.

Self Cite

View full text Add to dashboard Cite

To investigate the influence of miRNAs on target gene at transcription levels, we have systematically analyzed the expression profiles by a regression method. The results indicated that the down-regulation level of target mRNA is determined by multiple factors including the length and secondary structure of target mRNA, the structural accessibility of the target site and so on. Furthermore, by comparison of different miRNAs, the results indicated that, in animals, all miRNAs regulate their targets in a similar way.

show abstract

New syntax to describe local continuous structure-sequence information for recognizing new pre-miRNAs

Cited by 6 publications

References 24 publications

mirExplorer: Detecting microRNAs from genome and next generation sequencing data using the AdaBoost method with transition probability matrix and combined features

mirExplorer: Detecting microRNAs from genome and next generation sequencing data using the AdaBoost method with transition probability matrix and combined features

Finding Cancer-Associated miRNAs: Methods and Tools

Computational analysis on the level of miRNA-mediated target mRNA degradation

Contact Info

Product

Resources

About