Nucleosome Dynamics Help - Docker implementation
Nucleosome Dynamics - docker implementation
Nucleosome Dynamics docker image is a containerized implementation of 'Nucleosome Dynamics suite' particularly developed for use with the Galaxy platform.
Nucleosome Dynamics
Nucleosome Dynamics is a suite of R programs for nucleosome-related analyses based on MNase-seq experimental data. The toolkit includes the following analyses:
- nucleR: define and classify the location of nucleosomes from MNase-seq data.
- Nucleosome Dynamics: compares different MNase-seq experiments identifying variations in nucleosomes location.
- NFR: locate nucleosome-free regions
- Periodicity: predicts nucleosome phasing at gene level
- TSS: classify transcription start sites based on the surrounding nucleosomes
- Stiffness: computes nucleosome stiffness derived from a Gaussian function fitting
Container software
This docker image extends and distributes the following rellevant software:
- R libraries:
- NucleR R library: https://github.com/nucleosome-dynamics/nucleR
- Nucleosome Dynamics R library: https://github.com/nucleosome-dynamics/NucDyn
- R programs:
- Nucleosome Dynamics analyses: https://github.com/nucleosome-dynamics/nucleosome_dynamics
Build the image
The docker image for Nucleosome Dynamics can be found in the docker hub as mmbirb/nucleosome-dynamics
. However, this repository contains the data to manually rebuild it by running docker build.
docker build -t nucleosome-dynamics .
Note that the current working directory must contain the Dockerfile file.
Running nucleosome-dynamics container
You can either run the container manually, or via the web-based Galaxy platform. In both cases, individual nucleosome-related analyses are offered as:
- individual analyses
- workflow
Install docker for your system, if not previously done.
Running an individual analysis
You can run manually your containers using the following command:
docker run -v /path/to/data_dir/:/path/to/data_dir mmbirb/nucleosome-dynamics [analysis] [analysis_options]
In case you do not have the container stored locally, docker will download it for you. Once the execution is finished, the container will exit.
A short description of the parameters would be:
docker run
will run the container for you.-v /path/to/data_dir/:/path/to/data_dir
will make thedata_dir
where input files are stored available to the containermmbirb/nucleosome-dynamics
is the image name, which can be found in the docker hub.[analysis] [analysis_options]
correspond to the analysis type you want to run and its paremeters.
Available analysis are:
[analysis] |
Description |
---|---|
readBAM | Read Aligned MSase-seq BAM into a RData structure (required for further process) |
nucleR | Determine the positions of the nucleosomes across the genome |
NFR | Short regions depleted of nucleosomes |
txstart | Classify Transciption start accordint to the properties of surrounding nucleosomes |
periodicity | Periodic properties of nucleosomes inside gene bodies |
stiffness | Aparent stiffness constant foreach nucleosome obtained by fitting the coverage to a gaussian distribution |
nucleR_stats | Nucleosome call statistics |
nucDyn | Comparison of two diferent MNase-seq experiments to nucleosome architecture local changes |
NFR_stats | Nucleosome Free Regions statistics |
txstart_stats | TSS and TTS statistics |
periodicity_stats | Statistics on Nucleosome periodicity |
stiffness_stats | Statistics on stiffness |
nucDyn_stats | Statistics on Nucleosome Dynamics analysis |
Each analysis has its own input files and arguments. docker run mmbirb/nucleosome-dynamics [analysis] --help
will display such information in detail. Yet, a full usage description can be found at the Nucleosome Dynamics repository.
Example
Here is an example on how to load a MNase-seq reads file using readBAM
. It takes as input a BAM file and convert it into an RData file ready to be feed to other analyses ('nucleR', 'NFR', etc.).
docker run -v $PWD/test/data/:$PWD/test/data mmbirb/nucleosome-dynamics readBAM --input $PWD/test/data/cellcycleG2_chrII.bam --output $PWD/test/data/cellcycleG2_chrII.RData --type paired
Running a workflow script
You can combine different analysis tools to build your own workflow, you can do it at Galaxy platform or you can run it manually following this command:
docker run -v /path/to/data_dir/:/path/to/data_dir mmbirb/nucleosome-dynamics run [WF_file_path]
Where:
WF_file_path
: is a bash file containing the Rscript calls to different analyses.
Example
Here is an example on how to run the test workflow file test/scripts/wf-test.sh
, a bash file sequencially calling all 'Nucleosome Dynamics' analyses.
docker run -v /path/to/data_dir/:/path/to/data_dir mmbirb/nucleosome-dynamics run /path/to/data_dir/test/scripts/wf-test.sh