ncRNAScan: A pipeline to identify putative novel ncRNAs from deep sequencing data

ncRNAScan is a pipeline to extract putative novel ncRNAs ab initio, given a list of transcripts in GTF format assembled from deep sequencing data (ex: RNA-Seq) and annotation data.

This pipeline script will bind together the functionality of the tools / scripts: cuffcompare, categorize_ncRNAs.pl, get_unique_features.pl, fetch_seq_from_ucsc.pl, RNAfold, Infernal and Coding Potential Calculator (CPC.sh). Transcriptome construction tools such as Cufflinks produces a set of assembled transcripts in GTF format. ncRNAScan uses this data in addition to known gene annotation to extract putative ncRNAs constructed by the ab initio assemblers. The pipeline relies on the FPKM / RPKM values generated by these assemblers to assess the confidence of the constructed de novo transcripts and validates it against the known refenrece gene and non coding RNA information to identify putative novel ncRNAs.

IO::Routine

• The scripts use custom IO::Routine Perl Module.

• If you are installing ncRNAScan Pipeline, IO::Routine module is automatically installed.

☲☴ ncRNAScan

• Head on to NGS-Utils directory for script list.

• Install ncRNAScan and all its dependencies (Mac and Linux):

  cd /to/your/preferred/install/path curl -O https://raw.githubusercontent.com/biocoder/Perl-for-Bioinformatics/master/NGS-Utils/ncRNAScan perl ncRNAScan -setup

• Documentation:

  perl ncRNAScan -h

  or

  perldoc ncRNAScan

  or to get help documentation for individual modules, do:

  perl ncRNAScan -h cuff perl ncRNAScan -h cat perl ncRNAScan -h get perl ncRNAScan -h fetch perl ncRNAScan -h cpc perl ncRNAScan -h rna perl ncRNAScan -h inf

• Known issues:

  • If pipeline setup fails due to XML::Parser module, you need to install XML parser C libraries.

  • On Ubuntu / Debian based Linux distributions, as root user, do:

    apt-get install libexpat1 libexpat1-dev

  • On RedHat / Fedora / CentOS based Linux distributions, as root user do:

    yum install expat expat-devel
  • RNAfold: RNAfold is slow and does not work for sequences over 10000bp in length. I am working on including an alternative secondary structure prediction program instead of RNAfold. Meanwhile you may skip running RNAfold module by not issuing the --rnafold option with ncRNAScan.

• Caveats:

  • The pipeline script uses a lot of inherent Linux core utils and has been only tested in BASH shell.

Citation

Konganti, Kranti (2014). ncRNAScan: A pipeline to identify putative novel ncRNAs from deep sequencing data. ZENODO. 10.5281/zenodo.10566

Cheers,

BioCoder