Introduction
==============================================================================================================================
miREvo: an integrative microRNA evolutionary analysis platform for next-generation sequencing experiments
(version 1.3, Apr. 2015)
Please send bug reports, comments etc. to:
wenming126@gmail.com
This is miREvo developed by Wen Ming.
miREvo discovers active known or novel miRNAs from deep sequencing data (Solexa/Illumina, 454, ...),
This file explains the installation and Usage of miREvo on Linux systems.
Installation Instructions
==============================================================================================================================
To install miREvo type the following:
./Install.sh dir_for_install_miREvo
The parameter "dir_for_install_miREvo" is the directory to install miREvo, for example /home/user/miREvo
Configuration
==============================================================================================================================
Add the following line to you .bashrc file, which mostly places under your home directory: /home/user .
export MIREVO="/dir/for/install/miREvo"
And add MIREVO both to your environment variables PERL5LIB and PATH like this:
export PERL5LIB="$PERL5LIB:$MIREVO"
export PATH="$PATH:$MIREVO"
Dependencies
==============================================================================================================================
Linux system, 2GB Ram, enough disk space dependent on your deep sequencing data
Note most of the libraries and tools below can be easyly installed by apt-get if your system is ubuntu, which is the way we
recommended.
1) Perl (http://www.perl.org/)
2) Bioperl (http://www.bioperl.org/wiki/Main_Page)
3) Perl/Tk (http://search.cpan.org/~ni-s/Tk-804.027/pod/UserGuide.pod)
4) Blat (http://genome.ucsc.edu/)
5) Bowtie (http://bowtie-bio.sourceforge.net/index.shtml)
6) mafsInRegion (http://genomewiki.ucsc.edu/index.php/Kent_source_utilities)
7) RNAfold, RNAplot (http://www.tbi.univie.ac.at/~ivo/RNA/)
8) randfold (http://bioinformatics.psb.ugent.be/software/details/Randfold)
9) ImageMagick (http://www.imagemagick.org/script/index.php)
To test if everything is installed properly type in
1) bowtie
2) RNAfold -h
3) randfold
4) convert
you should not get any error messages otherwise something is not correctly installed
Note: some of those utilities are provided in the following package in case your don't want to build the whole source code for those softwares.
http://evolution.sysu.edu.cn/software/utilities.tar.gz
Getting started
==============================================================================================================================
miREvo can either operates via a Graphical User Interface (GUI) or from the command-line interface (CLI).
The options and parameters in the GUI are almost the samve as the (CLI).
The GUI also provides as a graphical interface to view many of the results and results can be more lively presented in
GUI. You can also use the GUI to view your the result run from command-line interface.
whether the project was assembled using the GUI or the CLI. Results appear as output files
using either the GUI or the CLI.
You can take a look at the command line parameters and available tools using
miREvo or miREvo_tk
Further information on each of the tools can be retrieved by the command pattern
miREvo <toolname>
for instance in the step of the filter sequencing reads
miREvo filter
Examples & Demos
==============================================================================================================================
We provide two sample project, which can be downloaded at
http://evolution.sysu.edu.cn/software/dme_demo.tar.gz or
http://evolution.sysu.edu.cn/software/ath_demo.tar.gz
These packages contains files for the two demos used in our miREvo software paper, Drosophila melanogaster and Arabidopsis thaliana.
Take the file dme_demo.tar.gz for example, this is a full demo for analysis of small RNA library generated from Drosophila melanogaster.
To run these project, you should have a Linux machine with at least 4 Gb RAM. This is mainly due to that miREvo (mafsInRegion) need to read
through the whole MAF file in ortholog extracting step.
To start the demo, do:
tar xzvf dme_demo.tar.gz
cd dme_demo
bowtie-build -f dme_repeat/database.fas dme_repeat/database
bowtie-build -f dme_mirbase/dme.hairpin.fa dme_mirbase/dme.hairpin
miREvo filter -i dme_reads.fa -d dme_repeat/database -H dme_mirbase/dme.hairpin -M dme_mirbase/dme.mature.fa -o dme -p 10
bowtie-build -f dme_chr/reference.fas dme_chr/reference
miREvo predict -o dme -r dme_chr/reference -M dme_mirbase/dme.mature.fa -s dme_mirbase/mature.nodme.fa -c -g 20000 -p 10
bowtie-build -f dme/predict.hairpin.fa dme/predict.hairpin
miREvo homoseq -i dme/filter.kn.nomap.fas -H dme/predict.hairpin -M dme/predict.mature.fa -r dme_chr/reference.fas -s dme_mirbase/mature.nodme.fa -m dm3.maf -o dme -p 10
To indentify the homology sequence of all known miRNA for Drosophila melanogaster , do
miREvo homoseq -i dme_reads.fa -H dme_mirbase/dme.hairpin -M dme_mirbase/dme.mature.fa -r dme_chr/reference.fas -m dm3.maf -s dme_mirbase/mature.nodme.fa -o kwn -p 10
For the Graphical User Interface. type:
cd dme_demo
miREvo_tk
1) Click button New to create a new project.
2) Click button Open to open a project previously run.
Then follow the instructions to go on you operation.
Input File Annotations:
-- dme_reads.fas: This is a small RNA reads library sequenced using Solexa 1G.
-- dme_repeat/database.fas: Fasta sequenced contains structure RNA (tRNA, transposon, etc.) or mRNA sequenced,
reads mapped to this sequences will be removed from the reads data set.
-- dme_mirbase/*.fa: The miRNAs of drosophila which previously identified, downloaded from miRBase.
-- dme_chr/*.fa: The genome sequence of D. melanogaster (dm3), downloaded from UCSC.
-- dm3.maf: Multiple alignments of 14 insects with D. melanogaster (dm3), downloaded from UCSC.
Output File Annotations:
See the following parts of this manual.
We also provide a simple solution to compare the expression of orthologous miRNAs across multiple libraries.
The is implemented as a perl script running as:
perl compare_exp.pl reference_genome homoseq_prj predict_prj1 predict_prj2 ...... predict_prjN
In the above command, compare_exp.pl is provided in utilities.tar.gz in our miREvo web site. reference_genome is the reference genome of one of your
interested species, which is also the reference genome of the MAF (WGAs) file; For example, the the reference_genome for tair8-11species.maf is
araTHA8. homoseq_prj is a project generated by "miREvo homoseq" command. predict_prj1, predict_prj2, ..., is the projects generated by "miREvo predict"
command and so on.
For more detailed information, see dro_orth_exp.readme (http://evolution.sysu.edu.cn/software/dro_orth_exp.readme) in miREvo web site.
Usage
==============================================================================================================================
filter
Description:
Small RNA reads are filtered out from the complex small RNA sequencing libraries following these two steps:
Firstly is the unwanted reads derived from any types of structure RNAs present in the cell, such as tRNA, rRNA annotated in Rfam;
RNA derived form repeat/transposon presented in RepBase; since most annotated miRNAs in miRBase are presented in intergenic region
reads corresponded to mRNA/coding sequencing can also be discarded.
Secondly is the reads represented for the known miRNA in the corresponding species, if it is provided. Their expression is also
determined in this step, using quantifier.pl provided in miRDeep2 package.
The remaining reads are then used for novel miRNA prediction.
Input:
1) Sequencing reads in the collapsed suffix fasta format.
Collapses reads in the fasta file to ensure that each sequence only occurs once. To indicate how
many times reads the sequence represents, a suffix is added to each fasta identifier. E.g. a
sequence that represents ten reads in the data will have the '_x10' suffix added to the identifier.
You can collapse reads with collapse_reads*.pl (collapse_reads_md.pl in miRDeep2-v2.0.0.7) provided by miRDeep2:
collapse_reads_md.pl reads.fa > reads_collapsed
2) Bowtie index database contains the unwanted sequence described above, for instance:
cat repeat.fas Rfam.contains.no.mir.fas mRNA.fas anyother.fas > database.fas
bowtie-built -f database.fas database
3) Bowtie index database for hairpin sequence for known miRNAs, please provided this if you only wanted
to predict novel miRNAs.
Example usage:
miREvo filter -i reads.fa -d /dir/to/database -H /dir/to/hairpin -o prj -p 10
Output:
All the output file will be dump to the "prj" folder (specified by the option "-o"), useful files are:
1) filter.statistic, basic statistic for the reads mappings and expression evaluation for known miRNAs.
2) filter.kn.map, reads distribution along each known miRNA.
3) filter.kn.nomap.fas reads
Notes:
In miREvo, the fasta sequences corresponded for the bowtie index may be also needed in some steps. For example,
if you provide bowtie index as "/dir/to/hairpin", miREvo will look for the corresponding fasta file in the format
as "/dir/to/hairpin.fa", "/dir/to/hairpin.fas" or "/dir/to/hairpin.fasta". Please name them in that format and
build the bowtie index in the right way.
predict
Description:
Perform a microRNA prediction by using deep sequencing reads.
Input:
1) filter.kn.nomap.fas, one of the outputs after the "filter" step.
2) Bowtie index database for genome sequence.
3) Fasta file with known miRNA mature sequence for your species.
Example usage:
miREvo predict -o prj -r /dir/to/genome -M /dir/to/mature.fas -s /dir/to/other.mature.fas -p 10 -m 1
Output:
1) predict.result.csv, miRDeep2 report prediction result.
2) predict.result.html, miRDeep2 report prediction result, html format.
3) predict.mirna.map, reads distribution along each predicted miRNA.
4) predict.hairpin.fa, hairpin sequences for predicted miRNAs;
5) predict.mature.fa, mature sequences for predicted miRNAs;
6) predict.statistic, basic statistic for reads mappings and expression evaluation for predicted miRNAs.
Notes:
Characterization of plant miRNAs are considerable different from animal miRNAs, for example, plant miRNAs may possess
a larger families and long hairpin sequence. By setting the right parameters for options "-k","-m" , miREvo will
tolerated more repeat mapping and will adjust to the parameters developed specifically for plant miRNA discovery.
homoseq
Description:
Fetching the homology alignment for miRNA sequence.
Input:
1) Fasta file with deep sequencing reads.
2) Fasta file with miRNA mature sequence.
3) Fasta file with miRNA hairpin sequence.
4) Bowtie index database for genome reference sequence.
5) Whole genome alignment files, UCSC MAF format.
Example usage:
miREvo homoseq -o prj -i /dir/to/reads.fa -M /dir/to/mature.fa -H /dir/to/hairpin.fa -d /dir/to/reference -m /dir/to/maf -p 10
Output:
1) homoseq.slim.aln, alignment of homology sequence for each miRNA.
2) homoseq.mirna.map reads distribution along each predicted miRNA.
3) homoseq.statistic, basic statistic for reads mappings and expression evaluation for predicted miRNAs.
4) homoseq.mirna.kmir, dna distance of miRNA hairpin (kmir), calculated based on Kimura 2-parameters model.
Notes:
1) For the command "homoseq" and "display" sequences searching of miRNA precursors.
The ids in files miRBase_mmu_v14.fa and precursors_ref_this_species.fa need to be similar to each other.
This is usually no problem if you downloaded both files from miRBase.
Otherwise it can happen that the quantifier fails to produce results.
2) MAF file must be constructed by using the genome of your species as the reference genome.
3) Please make sure that the chromosome name in the genome reference of your specie is the same as it present in the MAF file
Otherwise it can happen that miREvo fails to fetch the alignment.
display
Description:
Generating graphic report for structure and reads distribution information of miRNA.
Input:
1) Fasta file with deep sequencing reads.
2) Fasta file with miRNA mature sequence.
3) Fasta file with miRNA hairpin sequence.
Example usage:
miREvo display -o prj -i /dir/to/reads.fa -M /dir/to/mature.fa -H /dir/to/hairpin.fa -p 10
Output:
1) display.mirna.map reads distribution along each miRNA.
3) display.statistic, basic statistic for reads mappings and expression evaluation for miRNAs.
Notes:
For the "predict" step, the read mappings and structure information of the known miRNAs can not be browsed in the GUI.
You can use this command to create a new project and use the GUI command "open a existing project" to display
those information.
End
==============================================================================================================================
Thanks for using miREvo, have fun!
By: Wen Ming
Email: wenming126@gmail.com