Using a kernel density estimation based classifier to predict species-specific microRNA precursors

Tien-Hao Chang, Chih Ching Wang, Jian Wei Chen

Research output: Contribution to journalArticle

30 Citations (Scopus)

Abstract

Background: MicroRNAs (miRNAs) are short non-coding RNA molecules participating in post-transcriptional regulation of gene expression. There have been many efforts to discover miRNA precursors (pre-miRNAs) over the years. Recently, ab initio approaches obtain more attention because that they can discover species-specific pre-miRNAs. Most ab initio approaches proposed novel features to characterize RNA molecules. However, there were fewer discussions on the associated classification mechanism in a miRNA predictor. Results: This study focuses on the classification algorithm for miRNA prediction. We develop a novel ab initio method, miR-KDE, in which most of the features are collected from previous works. The classification mechanism in miR-KDE is the relaxed variable kernel density estimator (RVKDE) that we have recently proposed. When compared to the famous support vector machine (SVM), RVKDE exploits more local information of the training dataset. MiR-KDE is evaluated using a training set consisted of only human pre-miRNAs to predict a benchmark collected from 40 species. The experimental results show that miR-KDE delivers favorable performance in predicting human pre-miRNAs and has advantages for pre-miRNAs from the genera taxonomically distant to humans. Conclusion: We use a novel classifier of which the characteristic of exploiting local information is particularly suitable to predict species-specific pre-miRNAs. This study also provides a comprehensive analysis from the view of classification mechanism. The good performance of miR-KDE encourages more efforts on the classification methodology as well as the feature extraction in miRNA prediction.

Original languageEnglish
Article numberS2
JournalBMC Bioinformatics
Volume9
Issue numberSUPPL. 12
DOIs
Publication statusPublished - 2008 Dec 12

Fingerprint

Kernel Density Estimation
Spatial Analysis
MicroRNA
MicroRNAs
microRNA
Precursor
Classifiers
Classifier
Predict
seeds
RNA
Variable Kernel
Molecules
Kernel Density Estimator
Gene expression
Support vector machines
Feature extraction
Benchmarking
Untranslated RNA
Transcriptional Regulation

All Science Journal Classification (ASJC) codes

  • Biochemistry
  • Molecular Biology
  • Computer Science Applications
  • Structural Biology
  • Applied Mathematics
  • Agricultural and Biological Sciences (miscellaneous)
  • Artificial Intelligence
  • Computational Theory and Mathematics
  • Software

Cite this

@article{844fcd2fc1fe4d92845bbd9733162e95,
title = "Using a kernel density estimation based classifier to predict species-specific microRNA precursors",
abstract = "Background: MicroRNAs (miRNAs) are short non-coding RNA molecules participating in post-transcriptional regulation of gene expression. There have been many efforts to discover miRNA precursors (pre-miRNAs) over the years. Recently, ab initio approaches obtain more attention because that they can discover species-specific pre-miRNAs. Most ab initio approaches proposed novel features to characterize RNA molecules. However, there were fewer discussions on the associated classification mechanism in a miRNA predictor. Results: This study focuses on the classification algorithm for miRNA prediction. We develop a novel ab initio method, miR-KDE, in which most of the features are collected from previous works. The classification mechanism in miR-KDE is the relaxed variable kernel density estimator (RVKDE) that we have recently proposed. When compared to the famous support vector machine (SVM), RVKDE exploits more local information of the training dataset. MiR-KDE is evaluated using a training set consisted of only human pre-miRNAs to predict a benchmark collected from 40 species. The experimental results show that miR-KDE delivers favorable performance in predicting human pre-miRNAs and has advantages for pre-miRNAs from the genera taxonomically distant to humans. Conclusion: We use a novel classifier of which the characteristic of exploiting local information is particularly suitable to predict species-specific pre-miRNAs. This study also provides a comprehensive analysis from the view of classification mechanism. The good performance of miR-KDE encourages more efforts on the classification methodology as well as the feature extraction in miRNA prediction.",
author = "Tien-Hao Chang and Wang, {Chih Ching} and Chen, {Jian Wei}",
year = "2008",
month = "12",
day = "12",
doi = "10.1186/1471-2105-9-S12-S2",
language = "English",
volume = "9",
journal = "BMC Bioinformatics",
issn = "1471-2105",
publisher = "BioMed Central",
number = "SUPPL. 12",

}

Using a kernel density estimation based classifier to predict species-specific microRNA precursors. / Chang, Tien-Hao; Wang, Chih Ching; Chen, Jian Wei.

In: BMC Bioinformatics, Vol. 9, No. SUPPL. 12, S2, 12.12.2008.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Using a kernel density estimation based classifier to predict species-specific microRNA precursors

AU - Chang, Tien-Hao

AU - Wang, Chih Ching

AU - Chen, Jian Wei

PY - 2008/12/12

Y1 - 2008/12/12

N2 - Background: MicroRNAs (miRNAs) are short non-coding RNA molecules participating in post-transcriptional regulation of gene expression. There have been many efforts to discover miRNA precursors (pre-miRNAs) over the years. Recently, ab initio approaches obtain more attention because that they can discover species-specific pre-miRNAs. Most ab initio approaches proposed novel features to characterize RNA molecules. However, there were fewer discussions on the associated classification mechanism in a miRNA predictor. Results: This study focuses on the classification algorithm for miRNA prediction. We develop a novel ab initio method, miR-KDE, in which most of the features are collected from previous works. The classification mechanism in miR-KDE is the relaxed variable kernel density estimator (RVKDE) that we have recently proposed. When compared to the famous support vector machine (SVM), RVKDE exploits more local information of the training dataset. MiR-KDE is evaluated using a training set consisted of only human pre-miRNAs to predict a benchmark collected from 40 species. The experimental results show that miR-KDE delivers favorable performance in predicting human pre-miRNAs and has advantages for pre-miRNAs from the genera taxonomically distant to humans. Conclusion: We use a novel classifier of which the characteristic of exploiting local information is particularly suitable to predict species-specific pre-miRNAs. This study also provides a comprehensive analysis from the view of classification mechanism. The good performance of miR-KDE encourages more efforts on the classification methodology as well as the feature extraction in miRNA prediction.

AB - Background: MicroRNAs (miRNAs) are short non-coding RNA molecules participating in post-transcriptional regulation of gene expression. There have been many efforts to discover miRNA precursors (pre-miRNAs) over the years. Recently, ab initio approaches obtain more attention because that they can discover species-specific pre-miRNAs. Most ab initio approaches proposed novel features to characterize RNA molecules. However, there were fewer discussions on the associated classification mechanism in a miRNA predictor. Results: This study focuses on the classification algorithm for miRNA prediction. We develop a novel ab initio method, miR-KDE, in which most of the features are collected from previous works. The classification mechanism in miR-KDE is the relaxed variable kernel density estimator (RVKDE) that we have recently proposed. When compared to the famous support vector machine (SVM), RVKDE exploits more local information of the training dataset. MiR-KDE is evaluated using a training set consisted of only human pre-miRNAs to predict a benchmark collected from 40 species. The experimental results show that miR-KDE delivers favorable performance in predicting human pre-miRNAs and has advantages for pre-miRNAs from the genera taxonomically distant to humans. Conclusion: We use a novel classifier of which the characteristic of exploiting local information is particularly suitable to predict species-specific pre-miRNAs. This study also provides a comprehensive analysis from the view of classification mechanism. The good performance of miR-KDE encourages more efforts on the classification methodology as well as the feature extraction in miRNA prediction.

UR - http://www.scopus.com/inward/record.url?scp=57649207899&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=57649207899&partnerID=8YFLogxK

U2 - 10.1186/1471-2105-9-S12-S2

DO - 10.1186/1471-2105-9-S12-S2

M3 - Article

C2 - 19091019

AN - SCOPUS:57649207899

VL - 9

JO - BMC Bioinformatics

JF - BMC Bioinformatics

SN - 1471-2105

IS - SUPPL. 12

M1 - S2

ER -