RiboTaper
Detecting actively translated open reading frames in ribosome profiling data
Publication:
L.Calviello, N. Mukherjee, E. Wyler, H. Zauber, A. Hirsekorn, M. Selbach, M. Landthaler, B.Obermayer, U. Ohler(2016): Detecting actively translated open reading frames in ribosome profiling data Nature methods
About
RiboTaper is a new analysis pipeline for Ribosome Profiling (Ribo-seq) experiments, which exploits the triplet periodicity of ribosomal footprints to call translated regions.
The RiboTaper workflow consists of:
1) Creation of annotation files for of coding + noncoding genes.
2) Data tracks creation from Ribo-seq + RNA-seq experiments, dependent on P-sites calculation cutoffs.
3) Analysis of translation on exonic regions + quality control
4) de novo ORF finding in coding and non-coding regions
5) Creation of custom peptide fasta database + BED files for identified ORFs
Requirements
We ran the RiboTaper analysis on a SGE cluster, using 7 cores and h_vmem 8G. For each dataset the complete RiboTaper workflow (from the bam files to final results) took ~ 1 day
More investigation on the specific requirements for each dataset and usage on different machines will follow in the next future.
RiboTaper runs on Linux, it requires Samtools(0.1.19, must be in the PATH), Bedtools (v2.17.0) and R (3.0.1) with the following packages (available in the R_packages folder):
seqinr_3.1-3
ade4_1.7-2
multitaper_1.0-11
doMC_1.3.3
iterators_1.0.7
foreach_1.4.2
XNomial_1.0.1
To run the whole RiboTaper workflow you need (IMPORTANT: see the README file for a detailed description):
1) Genome fasta file
2) GTF file for annotation
3) BAM alignment files for Ribo-seq and RNA-seq
4) Read Lengths with respective cutoffs for P-sites calculation (for periodicity analysis), example: 29nt with +12 cutoff.
Files
A zip file containing the RiboTaper scripts, the commands we executed in our project, and a README file explaining the workflow in more details are available in the following links:
Version used in our project (outdated, 08.06.2015)
Version 1.2 (12.09.2015)
- Lower memory requirements
- Better plots for QC and final results.
- Support for different GTF files (TAIR10, latest ENSEMBL GTFs for Zebrafish and Rat)
- Final table with translated ORFs/genes + length/coverage per ORF category.
Current Version: 1.3 (recommended, 15.01.2016)
- Improved documentation
- Fixed bugs for "create_metaplots.bash" script.
- Improved P-site position calculation for a subset of spliced reads.
- More specific requirements for GTF file formatting
- Help with the use of StringTie/Cufflinks-derived GTF files.
Version: 1.3.1 (Beta Version, 20.04.2016)
- Stable GNU build system for configuration and installation
(Thanks to the help of Dilmurat Yusuf and Ricardo Wurmus, BIMSB).
Version: 1.3.1a (17.08.2016)
-Checks for the correct Rscript installation
-Checks for the correct bedtools version (<2.18)
(Thanks to the help of Dilmurat Yusuf and Ricardo Wurmus, BIMSB).
Thanks to Dilmurat Yusuf, the tool is also available via bioconda:
https://bioconda.github.io/recipes/ribotaper/README.html
and it is present in the Galaxy ToolShed:
Installation of conda package:
conda install ribotaper -c bioconda
Info regarding the Galaxy configuration:
1) At least 2 cores should be used for running ``ribotaper part 3: ribosome profiling``, therefore please configure job_conf.xml accordingly. For example, if local runner is used to run the job, the following example assigns two cores as the default,
<destination id="local" runner="local">
<param id="local_slots">2</param>
</destination>
2) Please add the following line to datatypes_conf.xml to enable the data type, compressed_archive,
<datatype extension="compressed_archive" type="galaxy.datatypes.binary:CompressedArchive" subclass="True" display_in_upload="True"/>
3) Please enable conda installation in galaxy.ini
conda_auto_install = True
conda_auto_init = True
R_packages + Bedtools executables (v2.17)
R_packages_Bedtools_2.17.tar.gz
Annotation files for Human and Zebrafish used in our project
Alignment files used in our project
Thank you for visiting the page and showing your interest in RiboTaper!
Do please contact me for any question!
Lorenzo.Calviello@mdc-berlin.de