A computational approach for identifying microRNA-target interactions using highthroughput CLIP and PAR-CLIP sequencing

Chih Hung Chou, Feng Mao Lin, Min Te Chou, S. heng Da Hsu, Tzu Hao Chang, Shun Long Weng, Sirjana Shrestha, Chiung Chih Hsiao, Jui Hung Hung, Hsien Da Huang

Research output: Contribution to journalArticle

28 Citations (Scopus)

Abstract

Background: MicroRNAs (miRNAs) play a critical role in down-regulating gene expression. By coupling with Argonaute family proteins, miRNAs bind to target sites on mRNAs and employ translational repression. A large amount of miRNA-target interactions (MTIs) have been identified by the crosslinking and immunoprecipitation (CLIP) and the photoactivatable-ribonucleoside-enhanced CLIP (PAR-CLIP) along with the next-generation sequencing (NGS). PAR-CLIP shows high efficiency of RNA co-immunoprecipitation, but it also lead to T to C conversion in miRNA-RNA-protein crosslinking regions. This artificial error obviously reduces the mappability of reads. However, a specific tool to analyze CLIP and PAR-CLIP data that takes T to C conversion into account is still in need. Results: We herein propose the first CLIP and PAR-CLIP sequencing analysis platform specifically for miRNA target analysis, namely miRTarCLIP. From scratch, it automatically removes adaptor sequences from raw reads, filters low quality reads, reverts C to T, aligns reads to 3'UTRs, scans for read clusters, identifies high confidence miRNA target sites, and provides annotations from external databases. With multi-threading techniques and our novel C to T reversion procedure, miRTarCLIP greatly reduces the running time comparing to conventional approaches. In addition, miRTarCLIP serves with a web-based interface to provide better user experiences in browsing and searching targets of interested miRNAs. To demonstrate the superior functionality of miRTarCLIP, we applied miRTarCLIP to two public available CLIP and PAR-CLIP sequencing datasets. miRTarCLIP not only shows comparable results to that of other existing tools in a much faster speed, but also reveals interesting features among these putative target sites. Specifically, we used miRTarCLIP to disclose that T to C conversion within position 1-7 and that within position 8-14 of miRNA target sites are significantly different (p value = 0.02), and even more significant when focusing on sites targeted by top 102 highly expressed miRNAs only (p value = 0.01). These results comply with previous findings and further suggest that combining miRNA expression and PAR-CLIP data can improve accuracy of the miRNA target prediction. Conclusion: To sum up, we devised a systematic approach for mining miRNA-target sites from CLIP-seq and PARCLIP sequencing data, and integrated the workflow with a graphical web-based browser, which provides a user friendly interface and detailed annotations of MTIs. We also showed through real-life examples that miRTarCLIP is a powerful tool for understanding miRNAs. Our integrated tool can be accessed online freely at http://miRTarCLIP. mbc.nctu.edu.tw.

Original languageEnglish
Article numberS2
JournalBMC Genomics
Volume14
DOIs
Publication statusPublished - Jan 21 2013

Fingerprint

Ribonucleosides
MicroRNAs
Immunoprecipitation
Argonaute Proteins
Web Browser
RNA
Workflow
3' Untranslated Regions

ASJC Scopus subject areas

  • Biotechnology
  • Genetics

Cite this

A computational approach for identifying microRNA-target interactions using highthroughput CLIP and PAR-CLIP sequencing. / Chou, Chih Hung; Lin, Feng Mao; Chou, Min Te; Hsu, S. heng Da; Chang, Tzu Hao; Weng, Shun Long; Shrestha, Sirjana; Hsiao, Chiung Chih; Hung, Jui Hung; Huang, Hsien Da.

In: BMC Genomics, Vol. 14, S2, 21.01.2013.

Research output: Contribution to journalArticle

Chou, Chih Hung ; Lin, Feng Mao ; Chou, Min Te ; Hsu, S. heng Da ; Chang, Tzu Hao ; Weng, Shun Long ; Shrestha, Sirjana ; Hsiao, Chiung Chih ; Hung, Jui Hung ; Huang, Hsien Da. / A computational approach for identifying microRNA-target interactions using highthroughput CLIP and PAR-CLIP sequencing. In: BMC Genomics. 2013 ; Vol. 14.
@article{0f164b6856ed4f4f9d832436711572d5,
title = "A computational approach for identifying microRNA-target interactions using highthroughput CLIP and PAR-CLIP sequencing",
abstract = "Background: MicroRNAs (miRNAs) play a critical role in down-regulating gene expression. By coupling with Argonaute family proteins, miRNAs bind to target sites on mRNAs and employ translational repression. A large amount of miRNA-target interactions (MTIs) have been identified by the crosslinking and immunoprecipitation (CLIP) and the photoactivatable-ribonucleoside-enhanced CLIP (PAR-CLIP) along with the next-generation sequencing (NGS). PAR-CLIP shows high efficiency of RNA co-immunoprecipitation, but it also lead to T to C conversion in miRNA-RNA-protein crosslinking regions. This artificial error obviously reduces the mappability of reads. However, a specific tool to analyze CLIP and PAR-CLIP data that takes T to C conversion into account is still in need. Results: We herein propose the first CLIP and PAR-CLIP sequencing analysis platform specifically for miRNA target analysis, namely miRTarCLIP. From scratch, it automatically removes adaptor sequences from raw reads, filters low quality reads, reverts C to T, aligns reads to 3'UTRs, scans for read clusters, identifies high confidence miRNA target sites, and provides annotations from external databases. With multi-threading techniques and our novel C to T reversion procedure, miRTarCLIP greatly reduces the running time comparing to conventional approaches. In addition, miRTarCLIP serves with a web-based interface to provide better user experiences in browsing and searching targets of interested miRNAs. To demonstrate the superior functionality of miRTarCLIP, we applied miRTarCLIP to two public available CLIP and PAR-CLIP sequencing datasets. miRTarCLIP not only shows comparable results to that of other existing tools in a much faster speed, but also reveals interesting features among these putative target sites. Specifically, we used miRTarCLIP to disclose that T to C conversion within position 1-7 and that within position 8-14 of miRNA target sites are significantly different (p value = 0.02), and even more significant when focusing on sites targeted by top 102 highly expressed miRNAs only (p value = 0.01). These results comply with previous findings and further suggest that combining miRNA expression and PAR-CLIP data can improve accuracy of the miRNA target prediction. Conclusion: To sum up, we devised a systematic approach for mining miRNA-target sites from CLIP-seq and PARCLIP sequencing data, and integrated the workflow with a graphical web-based browser, which provides a user friendly interface and detailed annotations of MTIs. We also showed through real-life examples that miRTarCLIP is a powerful tool for understanding miRNAs. Our integrated tool can be accessed online freely at http://miRTarCLIP. mbc.nctu.edu.tw.",
author = "Chou, {Chih Hung} and Lin, {Feng Mao} and Chou, {Min Te} and Hsu, {S. heng Da} and Chang, {Tzu Hao} and Weng, {Shun Long} and Sirjana Shrestha and Hsiao, {Chiung Chih} and Hung, {Jui Hung} and Huang, {Hsien Da}",
year = "2013",
month = "1",
day = "21",
doi = "10.1186/1471-2164-14-S1-S2",
language = "English",
volume = "14",
journal = "BMC Genomics",
issn = "1471-2164",
publisher = "BioMed Central Ltd.",

}

TY - JOUR

T1 - A computational approach for identifying microRNA-target interactions using highthroughput CLIP and PAR-CLIP sequencing

AU - Chou, Chih Hung

AU - Lin, Feng Mao

AU - Chou, Min Te

AU - Hsu, S. heng Da

AU - Chang, Tzu Hao

AU - Weng, Shun Long

AU - Shrestha, Sirjana

AU - Hsiao, Chiung Chih

AU - Hung, Jui Hung

AU - Huang, Hsien Da

PY - 2013/1/21

Y1 - 2013/1/21

N2 - Background: MicroRNAs (miRNAs) play a critical role in down-regulating gene expression. By coupling with Argonaute family proteins, miRNAs bind to target sites on mRNAs and employ translational repression. A large amount of miRNA-target interactions (MTIs) have been identified by the crosslinking and immunoprecipitation (CLIP) and the photoactivatable-ribonucleoside-enhanced CLIP (PAR-CLIP) along with the next-generation sequencing (NGS). PAR-CLIP shows high efficiency of RNA co-immunoprecipitation, but it also lead to T to C conversion in miRNA-RNA-protein crosslinking regions. This artificial error obviously reduces the mappability of reads. However, a specific tool to analyze CLIP and PAR-CLIP data that takes T to C conversion into account is still in need. Results: We herein propose the first CLIP and PAR-CLIP sequencing analysis platform specifically for miRNA target analysis, namely miRTarCLIP. From scratch, it automatically removes adaptor sequences from raw reads, filters low quality reads, reverts C to T, aligns reads to 3'UTRs, scans for read clusters, identifies high confidence miRNA target sites, and provides annotations from external databases. With multi-threading techniques and our novel C to T reversion procedure, miRTarCLIP greatly reduces the running time comparing to conventional approaches. In addition, miRTarCLIP serves with a web-based interface to provide better user experiences in browsing and searching targets of interested miRNAs. To demonstrate the superior functionality of miRTarCLIP, we applied miRTarCLIP to two public available CLIP and PAR-CLIP sequencing datasets. miRTarCLIP not only shows comparable results to that of other existing tools in a much faster speed, but also reveals interesting features among these putative target sites. Specifically, we used miRTarCLIP to disclose that T to C conversion within position 1-7 and that within position 8-14 of miRNA target sites are significantly different (p value = 0.02), and even more significant when focusing on sites targeted by top 102 highly expressed miRNAs only (p value = 0.01). These results comply with previous findings and further suggest that combining miRNA expression and PAR-CLIP data can improve accuracy of the miRNA target prediction. Conclusion: To sum up, we devised a systematic approach for mining miRNA-target sites from CLIP-seq and PARCLIP sequencing data, and integrated the workflow with a graphical web-based browser, which provides a user friendly interface and detailed annotations of MTIs. We also showed through real-life examples that miRTarCLIP is a powerful tool for understanding miRNAs. Our integrated tool can be accessed online freely at http://miRTarCLIP. mbc.nctu.edu.tw.

AB - Background: MicroRNAs (miRNAs) play a critical role in down-regulating gene expression. By coupling with Argonaute family proteins, miRNAs bind to target sites on mRNAs and employ translational repression. A large amount of miRNA-target interactions (MTIs) have been identified by the crosslinking and immunoprecipitation (CLIP) and the photoactivatable-ribonucleoside-enhanced CLIP (PAR-CLIP) along with the next-generation sequencing (NGS). PAR-CLIP shows high efficiency of RNA co-immunoprecipitation, but it also lead to T to C conversion in miRNA-RNA-protein crosslinking regions. This artificial error obviously reduces the mappability of reads. However, a specific tool to analyze CLIP and PAR-CLIP data that takes T to C conversion into account is still in need. Results: We herein propose the first CLIP and PAR-CLIP sequencing analysis platform specifically for miRNA target analysis, namely miRTarCLIP. From scratch, it automatically removes adaptor sequences from raw reads, filters low quality reads, reverts C to T, aligns reads to 3'UTRs, scans for read clusters, identifies high confidence miRNA target sites, and provides annotations from external databases. With multi-threading techniques and our novel C to T reversion procedure, miRTarCLIP greatly reduces the running time comparing to conventional approaches. In addition, miRTarCLIP serves with a web-based interface to provide better user experiences in browsing and searching targets of interested miRNAs. To demonstrate the superior functionality of miRTarCLIP, we applied miRTarCLIP to two public available CLIP and PAR-CLIP sequencing datasets. miRTarCLIP not only shows comparable results to that of other existing tools in a much faster speed, but also reveals interesting features among these putative target sites. Specifically, we used miRTarCLIP to disclose that T to C conversion within position 1-7 and that within position 8-14 of miRNA target sites are significantly different (p value = 0.02), and even more significant when focusing on sites targeted by top 102 highly expressed miRNAs only (p value = 0.01). These results comply with previous findings and further suggest that combining miRNA expression and PAR-CLIP data can improve accuracy of the miRNA target prediction. Conclusion: To sum up, we devised a systematic approach for mining miRNA-target sites from CLIP-seq and PARCLIP sequencing data, and integrated the workflow with a graphical web-based browser, which provides a user friendly interface and detailed annotations of MTIs. We also showed through real-life examples that miRTarCLIP is a powerful tool for understanding miRNAs. Our integrated tool can be accessed online freely at http://miRTarCLIP. mbc.nctu.edu.tw.

UR - http://www.scopus.com/inward/record.url?scp=84920574625&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84920574625&partnerID=8YFLogxK

U2 - 10.1186/1471-2164-14-S1-S2

DO - 10.1186/1471-2164-14-S1-S2

M3 - Article

AN - SCOPUS:84920574625

VL - 14

JO - BMC Genomics

JF - BMC Genomics

SN - 1471-2164

M1 - S2

ER -