run_analysis.md

Running the analysis

Make sure you have installed required software, packages and downloaded necessary files.

If you want to replicate the analysis as published, navigate to the phyloRNAanalysis folder and type:

Rscript run.r

This will remap BAM files to reference genome, detect expression levels and SNV, prepare FASTAs and perform phylogenetic reconstruction.

Running on your own data

If you want to use this for your own data, put your data into the data directory and delete the old data. Then modify the chemistry, densities, hdi and selection to suit your needs.

• chemistry -- While the automatic detection of chemistry is preferred, cellranger might fail to detect chemistry for low-quality data, for this reason, the chemistry was fixed in the analysis. Either change it to auto or to chemistry of your data.
• densities -- Filter data to a particular data density, set to a single value if you do not care about comparing phylogenies from different data densities.
• hdi -- Highest Density Interval method for discretization of expression values. This method will only work for homogeneous population of cells. If your population has heterogeneous expression levels, this method of discretization will not work.
• selection -- Named vector to select the best performing cells from each sample for the alternative filtering method used in the study.