Output Files

Output Files

Here, we assume you've run infercnv using the 2-step method involving 'infercnv::run()'. An output directory is specified in the run() call, and will contain several useful outputs as described below.

Note, the current inferCNV software generates a lot of output files, and only a small subset are of greatest interest. In the future, outputs will be better organized to facilitate identification of the most useful outputs.

• infercnv.preliminary.png : the preliminary inferCNV view (prior to denoising or HMM prediction)

• infercnv.png : the final heatmap generated by inferCNV with denoising methods applied.

• infercnv.references.txt : the 'normal' cell matrix data values.

• infercnv.observations.txt : the tumor cell matrix data values

• infercnv.observation_groupings.txt : group memberships for the tumor cells as clustered.

• infercnv.observations_dendrogram.txt : the newick formatted dendrogram for the tumor cells that matches the heatmap.

inferCNV object files

During the inferCNV data processing, it will write persistent object files representing the inferCNV object state at the corresponding stage of the processing. Each processing stage is numbered, so it's straightforward to follow the operations performed.

These objects can be resurrected in an R session like so:

library(infercnv)
infercnv_obj = readRDS('preliminary.infercnv_obj')

The above 'preliminary' object is used to generate the 'infercnv.preliminary.png' file, and the data included are the target for either denoising or performing CNV prediction using HMMs.

If you want to interact with any of the object files, the essential features to be aware of include the following slots of the inferCNV S4 object:

• 'infercnv_obj@expr.data' : contains the processed expression matrix as it exists at the end of that stage for which that inferCNV object represents.

• 'infercnv_obj@reference_grouped_cell_indices' : list containing the expression matrix column indices that correspond to each of the normal (reference) cell types.

• 'infercnv_obj@observation_grouped_cell_indices' : similar list as above, but corresponds to the tumor cell types.

Based on the above slots, it would be straightforward to extract info of interest and/or move data into other analysis frameworks.