MixviR is a tool to help analyze, explore, and visualize high-throughput genomic data from samples that may contain mixed lineages of a given microbial taxon. It was originally developed to analyze SARS-CoV-2 environmental (wastewater) samples, but can be applied to samples from any microbial taxon. It uses vcf-formatted input files along with reference genomic information to identify genetic variants (both at the DNA and amino acid level), and if provided optional input information about lineage-characteristic mutations, can estimate frequencies of target lineages within samples.


MixviR is available in R from CRAN with… install.packages(“MixviR”)

It can also be optionally downloaded and installed from github (https://github.com/mikesovic/MixviR) with… devtools::install_github(“mikesovic/MixviR/MixviR_X.Y.Z”), after replacing X.Y.Z with the current version.


Most MixviR analyses will start by running the call_mutations() function, which requires a directory storing one or more vcf files, a fasta-formatted reference genome file, and an annotation file associated with the reference genome that defines genes/open reading frames. If analyzing SARS-CoV-2, a pre-consructed reference object is available (based on the Wuhan reference) and can be specified with reference = “Wuhan”, meaning the sample vcf(s) are the only required input.

The call_mutations() function produces a data frame that can be saved as an object and/or written to a file, and that is used as input for the explore_mutations() and estimate_lineages() functions.

Getting Help

Aside from the vignette associated with the program, we encourage you to use the MixviR Google Group to get help with the program.