VISTA (Vaginal Inference of Subspecies and Typing Algorithm) mgCST and mgSs reference models

Data and code to run the VISTA and the app: https://github.com/JHolm-Lab/VISTA
Citation: Williams A, Maros A, France MT, Ravel J, Holm JB.2026. Not all vaginal microbiomes are equal: functional context shapes immune landscapes. mBio 17:e03645-25. https://doi.org/10.1128/mbio.03645-25

Unlike other parts of the body, a healthy vaginal microbiome is usually simple, with just a few types of bacteria—mostly Lactobacillus species. When the vaginal microbiome becomes more diverse and includes bacteria like Gardnerella, Sneathia, Prevotella, or “Candidatus Lachnocurva vaginae”, it’s considered less optimal and can increase the risk of reproductive and urinary health problems.

To describe these differences, scientists often use a system called Community State Types (CSTs), which groups vaginal microbiomes based on the types and amounts of lactic acid–producing bacteria they contain. For example:
CST I: mostly Lactobacillus crispatusCST II: mostly Lactobacillus gasseriCST III: mostly Lactobacillus inersCST IV: mostly diverse anaerobic bacteriaCST V: mostly Lactobacillus jenseniiHowever, this system doesn’t capture the full complexity. Many strains and species in the vaginal microbiome go undetected with traditional methods. These hidden differences can be important for understanding how the microbiome affects health.
To dig deeper, researchers look at the genetic makeup of the bacteria—specifically, the full set of genes found in each species. These genetic groupings are called metagenomic subspecies (mgSs). When you combine all the mgSs in a sample, you get a metagenomic Community State Type (mgCST), which gives a more detailed picture of how the microbiome might be influencing health.

This software package, VISTA (Vaginal Inference of Subspecies and Typing Algorithm), allows for classification of vaginal metagenomes to mgCSTs.