Systematic and computational identification of Androctonus crassicauda long non-coding RNAs

The potential function of long non-coding RNAs in regulating neighbor protein-coding genes has attracted scientists' attention. Despite the important role of lncRNAs in biological processes, a limited number of studies focus on non-model animal lncRNAs. In this study, we used a stringent step-by-step filtering pipeline and machine learning-based tools to identify the specific Androctonus crassicauda lncRNAs and analyze the features of predicted scorpion lncRNAs. 13,401 lncRNAs were detected using pipeline in A. crassicauda transcriptome. The blast results indicated that the majority of these lncRNAs sequences (12,642) have no identifiable orthologs even in closely related species and those considered as novel lncRNAs. Compared to lncRNA prediction tools indicated that our pipeline is a helpful approach to distinguish protein-coding and non-coding transcripts from RNA sequencing data of species without reference genomes. Moreover, analyzing lncRNA characteristics in A. crassicauda uncovered that lower protein-coding potential, lower GC content, shorter transcript length, and less number of isoform per gene are outstanding features of A. crassicauda lncRNAs transcripts.

Automated summary

The study identified specific lncRNAs in Androctonus crassicauda and analyzed their characteristics using a stringent filtering pipeline and machine learning tools. The majority of these lncRNAs were found to be novel with no identifiable orthologs in closely related species. The analysis revealed unique features of A. crassicauda lncRNAs, including lower protein-coding potential, lower GC content, shorter transcript length, and less number of isoform per gene.