Introduction

Numerous experimental and computational researches have expanded a number of diverse RNA-RNA interactions (RRIs). However, there are few text mining systems for extracting diverse RRIs information from biomedical literatures. Hence, we propose a text mining software, RIscoper (RNA Interactome Scoper), to extract RRIs. Notably, a reliable RRI corpus was integrated in RIscoper, recruiting more than 13,300 manually curated sentences with RRI information. RIscoper supports users to upload full texts or abstracts, as well as provides an online search tool connection with PubMed (PMID and keyword input), which are useful for biologists. In evaluation, RIscoper presents a high precision performance (90.4% precision and 93.9% recall) with integrating natural language processing techniques and reliable RRI corpus.


Highlights

  • RIscoper is based on N-gram statistics language model that is the first tool for full-scale RNA interactome scanning, which supports users to upload full texts or abstracts, as well as provides an online search tool connection with PubMed (PMID and keyword input).
  • RIscoper establishes a comprehensive and reliable RRI corpus, recruiting 13,377 sentences with RRI information which have been manually curated from more than 5,000 biomedical literatures. These positive sentences involved in multiple RNA interactions including mRNA, lncRNA, miRNA, sRNA, circRNA, snoRNA, snRNA, scaRNA and scRNA. It providing a favorable resource for ongoing text mining studies of RRIs and will be a benchmark dataset in other future machine learning works.
  • RIscoper is a simple and practical tool for database curators, experimental biologists as well as bioinformaticians.


News


  • RIscoper completed
    September 2018

  • RIscoper construction
    September 2017

  • Data collection
    April 2017




Contact


Wang Dong



wangdong@ems.hrbmu .edu.cn