This is an old repository. Please refer to https://github.com/Gaius-Augustus/VARUS for an updated version.
VARUS automates the selection and download of RNA-Seq runs available at the SRA, with regard to a sufficiently high coverage. This is done in a stepwise procedure. An iteration includes:
- selecting a run to download, that is expected to complement previously downloaded reads the most
- download the run with fastq-dump
- align the reads with STAR
- evaluate the alignment
The key here is that runs are only downloaded partially in each iteration. With these read-samples, estimations of the runs value for further downloads from this run are made. This allows VARUS to distinguish good runs from bad runs within the first few downloads, and download more extensively from runs that are likely to contain more reads from yet underrepresented transcripts.
Invoke the following command from the command-line in order to clone the repository:
git clone https://github.com/WillyBruhn/VARUS.gitVARUS incorporates the tools fastq-dump [https://ncbi.github.io/sra-tools/fastq-dump.html] and STAR [https://github.com/alexdobin/STAR]. You can install both tools and compile VARUS by invoking:
VARUS/./install.shIf you choose to install fastq-dump and STAR manually don't forget to also compile VARUS manually by invoking:
cd Implementation
makeFind a working setup under /GettingStarted/Pombe. Change to this folder and invoke the following command to start downloading runs:
../.././runVARUS.plThe final output is a file called VARUS.bam.
You can use this for your applications by modifying VARUSparameters.txt and species.txt and substituting pombe.fasta with a corresponding genome-file.
Pay attention to the following parameters:
--pathToSTAR must point to the executable of STAR.
--maxBatches specifies how many runs you want to download at most.
--batchSize specifies how many reads you want to download in each step.
Find the bachelor thesis corresponding to VARUS in /docs/Thesis.