Computational Approaches in Identifying Long Non-coding RNA

Long non-coding RNA (lncRNA) is the largest non-protein and functional RNA. The majority of lncRNAs are functionally uncharacterized. Thus, researchers are employing both experimental as well as computational approaches to characterize unknown lncRNAs. Recently, the majority of lncRNAs have been characterized using transcriptome sequencing datasets under different conditions. The information on these transcripts was mainly restricted to genomic loci and expression patterns. This discrepancy in lncRNAs’ functional understanding has largely been due to the lack of methods to classify basic genome-scale bio-molecular interactions and resources that systematically archive these interactions. Additionally, experimental methods are time-consuming and cost-effective. To overcome this, various bioinformatics tools have been developed to predict lncRNA. Thus, in this chapter, the authors attempted to understand the underlying principle of various computational approaches to detect lnRNAs. Information obtained reveals that lncRNA may be annotated by employing its coding potential and sequence conservation, folding algorithms, and interactions. Additionally, several databases have also been developed to help researchers detect lncRNAs from the large genomic dataset. Though the result obtained from these tools and databases is useful, a systematic integrative and metadata analysis is also required to understand diverse lncRNA regulatory mechanisms of action at various levels. Further attempts would be made to annotate their features, which is beneficial in understanding the underlying cell biology.