This page describes how to install a set of Perl scripts designed to identify members of known riboswitch classes in genomic sequences. It accompanies a Methods in Molecular Biology protocol chapter that explains how to use these scripts. This protocol may also be useful for analyzing other structured RNAs, such as those found in the Rfam database, or for investigating new regulatory RNA motifs, such as might be predicted by CMfinder.
Please note: Infernal and BioPerl have changed since this protocol was published, so the directions here differ somewhat from those in the book chapter. If you have any feedback, please let me know. You can also visit the original installation instructions (v1). |
Infernal is used to conduct covariance model searches. Download, compile, and install according to the directions for your system. There have been some interface changes since v0.81. This updated protocol assumes you have v1.0.
As of August 2009, the current stable release of BioPerl (v1.6) does not include the necessary modules for parsing Infernal search results or manipulating Stockholm alignments. Therefore, you need to download and install the current repository version of bioperl-live. Follow the Instructions for downloading any BioPerl repository using anonymous SVN.
libGD is required to produce drawings of the genomic contexts of putative RNA structure matches. Download and install according to the directions for your system.
Several additional Perl modules from are required for the riboswitch perl scripts. You can download and install them directly from CPAN. Alternately, on MacOS X, you can use the following command sequence if you have administrator privileges:
>sudo perl -MCPAN -e shell
>Password: ********
>install Storable
>install GD
>install Bio::Graphics
The GD module requires libGD to be installed beforehand and is only necessary for producing genomic context drawings. If you encounter any other messages of the form: "Can't locate module.pm
in @INC", try installing module
from CPAN.
Download the riboswitch Perl scripts.
![]() |
version 2 | 23 June 2008 |
Each Perl script has POD documentation. To view it, either open the script in a text editor or use one of these commands:
>perldoc genome_context.pl
>genome_context.pl --man
With the new version of Infernal (v1.0), the command sequence for performing global alignment of the riboswitch model to matches in the database with the hmmfilter has changed as follows:
>cmbuild TPP.cm TPP.sto
>cmsearch -g -T 27.2 TPP.cm NC_009832.1.fna > hits.txt
>cmsearch_reformat.pl -s scores.tab hits.txt hits.fna
>cmalign -o hits.sto TPP.cm hits.fna
>stockholm_to_html.pl hits.sto hits.html
>genomic_context.pl -b scores.tab -s NC_009832.1.gbk hits.sto context
With the new version of Infernal (v1.0), the example command sequence for performing global alignment of the riboswitch model to matches in the database has changed as follows:
>fragment_db.pl -s 40 /db/refseq24/fasta /db/refseq24/fragment
>cmbuild TPP.cm TPP.sto
>cmsearch -g TPP.cm /db/refseq24/fragment/1.fna > hits.1.txt
>cat hits.* > hits.txt
>cmsearch_reformat.pl -T 10 -b scores.tab hits.txt hits.fna
>cmalign -o hits.sto TPP.cm hits.fna
>stockholm_to_html.pl hits.sto hits.html
>index_db.pl -d refseq24_genbank /db/refseq24/genbank /db/index
>genomic_context.pl -b scores.tab -d /db/index/refseq24 hits.sto context
Barrick Lab > SupplementList > SupplementRiboswitchPrediction