Universal Genome Analyst (uga) is a tool designed to assist biomedical researchers in complex genomic data analysis
Universal Genome Analyst
Universal Genome Analyst (uga) is an open, flexible, and efficient tool for the distribution, management, and visualization of whole genome data analyses. It is designed to assist biomedical researchers in complex genomic data analysis through the use of a low level interface between the powerful R statistical environment and Python, allowing for rapid integration of emerging analytical strategies. This project uses Cython for a significant reduction in computation time and researchers with access to a high performance computing cluster or with access to multiple cores will find time-saving features for parallel analysis using a flexible, yet controlled, commandline interface.
This software is currently under rapid development. Updates and bug fixes are being tracked on the uga github page
- Notable Features
- Single variant association modeling (R base: lm, glm; R geepack: geeglm, R seqMeta: singlesnpMeta, R lme4: lmer)
- Gene/Group based association modeling (with meta analysis: R seqMeta: burdenMeta, skatMeta, skatOMeta)
- Family based single variant association modeling
- Publication quality Q-Q and manhattan plots
- Genomic control correction
- Post modeling meta analysis
- Run multiple models as a single submission (variant names need not match)
- Alignment of compatible variants based on genomic position and both alleles (A/T and G/C SNVs are ambiguous and are assumed to be pre-aligned)
- Automatic job splitting (with job array queueing)
- Input data split by chromosome can be linked via wildcard
- Automatically submit jobs on parallel computing systems using qsub
- multiple processor parallelization in addition to cluster parallelization
- User definable buffered reading for RAM usage control
- Verification and compilation for parallel distributed jobs
- Gzip and Bgzip / Tabix mapped output where possible to save disc space
- Planned For Future Releases
Since parallel computing is sometimes unreliable, analysts need to be able to verify and possibly rerun failed jobs at various stages of the analysis. In the interest of user efficiency and to avoid limitations induced by excessive automation, uga breaks the analytical process into the following modules.
- settings user definable settings
- snv single variant statistical modeling
- snvgroup gene/region-based statistical modeling
- meta meta-analysis
- compile verify and compile split analysis results
- resubmit automatically resubmit failed jobs for a project
- snvplot Q-Q and manhattan plots for snv tests
- snvgroupplot Q-Q and manhattan plots for snvgroup tests
- filter filter results / apply genomic control to results
- merge merge and annotate results with external files
- tools run any command line tool with ability to include genomic region automatically
This software uses a variety of Python modules, R packages, and some stand-alone software. Thus, the easiest method for installation is to use one of two platforms of the software conda; either Anaconda or Miniconda.
To prepare your system for uga, you need to clone an environment. You will need the included environment.yml file from the source code and a number of packages from my anaconda cloud channel and other custom channels (listed in the environment.yml file). After downloading the most recent release (available here), use the following commands to begin the installation.
For the sake of this tutorial, let’s assume the release version is ‘X’.
>>> tar -xvf uga-X.tar.gz >>> cd uga-X
At this point you may change the name of the environment to anything you’d prefer by modifying the first line of the environment.yml file. For these instructions, we will assume the name is unchanged from ‘uga’.
>>> conda env create -f environment.yml >>> source activate uga
Now that your environment is activated, you are ready to install uga from source.
>>> python setup.py install
Cutting Edge Install
Keeping up with the most current changes may be of interest to you as I will likely continue to add features and fix bugs on a regular basis. Thus, you may want to run a fork of this repository rather than installing from source. See a tutorial describing how to fork this repository.
If you install uga under a conda environment, you need to source the environment as shown above before running any task in uga.
>>> source activate uga
Verify that uga is functional using the following command to display help.
>>> uga -h
Note: further help is provided after selecting a specific module, ie.
>>> uga snv -h
While you may simply run uga on a single cpu system, if you have access to a parallel computing cluster or even a single multiple core processor, you will be able to take advantage of the self-managed parallel mode of use for which this software was designed. This release was tested on a system which deploys Sun Grid Engine and qsub for job management and will likely be compatible with other PBS systems.
Please cite this software as follows. A manuscript is in the works and yet to be submitted.
Koesterer, Ryan. Universal Genome Analyst (uga). https://github.com/rmkoesterer/uga. DOI: 10.5281/zenodo.578712.
- Author: Ryan Koesterer
Please report any bugs or issues using the Issues tab on this page. I will respond to all concerns as quickly as possible.
Universal Genome Analyst (uga) is distributed under the GNU General Public License v3:
Copyright (c) 2015 Ryan Koesterer
This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with this program. If not, see <http://www.gnu.org/licenses/>