automated system for extracted RRIs from literatures


Fully comprehensive views of diverse RNA-RNA interactions (RRIs) are essential for a system-level understanding of cellular behavior. Numerous experimental and computational researches have expanded a number of diverse RNA-RNA interactions which are not systematically archived and scattered. However, there are no text mining systems for extracting diverse RRIs information from biomedical literatures. To complement with this absence, we propose RIscoper to extract RRI from biological and biomedical text.


(1). RIscoper support user-provided abstracts or full text papers, such as PDF and TXT file extraction, and network communication with PubMed by online PubMed ID and keyword-based extraction.
(2). This study firstly establish a comprehensive and reliable RRI corpus, recruiting more than 13300 sentences with RRI information by manually curated. It’s providing a favourable resource for ongoing text mining studies of RNA interactions
(3). RIscoper is based on N-gram statistics language model, N-gram model has the following advantages:
      (i) has lower computational complexity;
      (ii) do less manual intervention of the sentences;
      (iii) can be easily transplanted and extended.
(4). RIscoper is a user friendly software written in JAVA, which is a fast and simple tool for database curators, experimental biologists as well as bioinformaticians.