Running eukrhythmic with the sample data

Note that running eukrhythmic on the sample data will take a little while. When we tested this functionality on our system, we used the following amounts of time and memory:

The sample data being run can be found in input/testdata/, and contains five small sample raw read files. The sample metaT file for this dataset is in input/sampledata.txt:

SampleName  SampleID        AssemblyGroup   FastqFile
HN008_subsampled    HN008   samplegroup1    HN008/HN008_subsampled
HN009_subsampled    HN009   samplegroup2    HN009_subsampled
HN016_subsampled    HN016   samplegroup1    HN016_subsampled
HN036_subsampled    HN036   samplegroup2    HN036_subsampled
HN043_subsampled    HN043   samplegroup2    HN043_subsampled

In the SampleName column, we put our longer descriptive name that matches our filename, whereas SampleID contains the smallest unique token we can make out of our sample names. In AssemblyGroups, we list the files we want the assembly software to have to assemble simultaneously. Even though all of our samples will eventually get combined, some of them will be treated independently to begin with by the assembly tools, and others won’t. In this example, HN008 and HN016 are likely from the same site, for example.

In FastqFile, we use the full path of the file relative to the sample directory (which in our case is input/testdata). HN008 has a subdirectory, so this syntax lets us leave our file organization as it is as we’re running eukrhythmic.

To run eukrhythmic on the provided sample data, invoke eukrhythmic on a clean install of the program without arguments. You can also run the sample data by using the argument --use-sample, which will copy the relevant configuration entries.

Without any additional flags, eukrhythmic will be run against the provided sample data on your local machine (or your current node on the cluster, if you’re logged into one). You should see a dialogue like this one: