An interactive RShiny app for exploring the variation across 16S rRNA gene copies in microbial taxa, helping researchers choose the right variable regions for their amplicon sequencing experiments.
Bacterial species carry multiple copies of the 16S rRNA gene (~5 on average, up to 15+), and those copies are often not identical. When planning an amplicon sequencing experiment, the variable region (v-region) you choose to sequence determines whether you can actually distinguish the species you care about from closely related taxa.
RADx (Regional Alignment Database explorer) lets you select any set of species and immediately visualize how their 16S rRNA gene copies differ across all nine v-regions. You can easily see which v-regions (V1-V9) produce unique sequences for your target taxa, which ones leave species indistinguishable from one another, and whether a single region is enough for your experimental needs.
This matters more than most researchers expect. The most commonly used v-regions in amplicon sequencing, V3 and V4, are frequently insufficient to differentiate between closely related organisms within the same genus or species group.
RADx is built on two reference libraries:
- RADlib16S - A bacteria and archaea 16S rRNA database containing all copies of the 16S rRNA gene in an organism
- RADlibVR - A database of all variable regions found in RADlib
When you select species in RADx, RADalign runs multiple sequence alignment and grouping analyses on those sequences across whichever variable regions you choose. RADx then visualizes the results so you can interpret them interactively.
RADexplorer is written and run in R. If you do not already have R downloaded, you can download it here, along with RStudio, the IDE most commonly used for coding R: https://posit.co/downloads/
RADx is installed directly from GitHub. All dependencies are handled automatically.
Run the following commands in your R or RStudio console:
Step 1: Install pak (if not already installed)
install.packages("pak")Step 2: Install Biostrings
install.packages("BiocManager")
BiocManager::install("msa")Step 3: Install RADalign
pak::pak("RADSuite/RADalign")Step 4: Install RADx
pak::pak("RADSuite/RADexplorer")Step 5: Load and launch
library(RADexplorer)
RADexplorer::run_app()When the app opens you will see the RADx menu. Use the species picker to search for and select the taxa you want to analyze. You can select as many species as you need. Keep in mind, the more species you select, the longer it will take to load the explorer. To select all species within a genus, use the "All Species (#)" options.
After the explorer loads, you can then select the v-regions you want to explore in the sidebar. By default V4 is selected, since it is the most commonly sequenced region, but you can select any combination of V1 through V9.
The main plot shows the selected species on the y axis and the v-regions on the x axis. In the default summarized view, all gene copy sequences within a v-region are merged into one combined 'signature'. A unique signature in a v-region means the organism can be identified by sequencing that region alone. Signatures are represented by tile color.
- A gold highlight on a column marks the currently selected variable regions
- A green checkmark next to a species means it can be uniquely identified using the selected variable regions
- A red bracket groups species that cannot be distinguished from one another with the selected regions
Note: Colors designate identical signatures within a v-region. Colors should not be compared across columns.
Toggle Detailed View in the sidebar to switch from a summary view to a copy-level view, showing each individual gene copy as its own row per species. In general, the summary view is best for interspecies comparison and v-region selection, while the detailed view is best for studying intraspecies copy variation within a taxon.
Note: Colors designate identical sequences within a v-region. Colors should not be compared across columns.
Toggle V-Region Labels to display sequence ID labels directly on each tile for closer inspection and discernment between v-region sequences.
Use the Locate Taxa picker to search for specific taxa in the plot. Located taxa will be marked with a blue arrow on the y axis and the plot will be filtered to include only the taxa they are are grouped with under the current v-region selection.
1) "Can I distinguish two closely related species with V4 alone?"
Select both species, check only V4, and see whether they fall into the same group (red bracket) or are identifiable (green checkmark).
2) "How many 16S copies does my target organism carry, and are they identical?"
Switch to Detailed View to see every individual gene copy per species. Tiles of different colors in the same column indicate unique sequences for that region.
3) "My budget only allows for sequencing one variable region. Which should I choose?"
Select your taxa and step through each V-region one at a time. The region that gives you the most green checkmarks across your taxa of interest is your best option.
4) "Another researcher used V3 amplicon sequencing. Can I compare my V4 data to theirs?"
Select your taxa and check V3 and V4 separately. If the groupings and checkmarks change between the two selections, the two datasets are identifying your community differently and direct comparison may be difficult.
5) "I am writing a methods section and need to justify the variable region I chose to sequence."
Select your taxa and check your chosen region. RADx shows you exactly why that region identifies your specific targets. Feel free to cite RADx as justification for your primer selection.
6) "I think two (or more) species in my data are collapsing into one OTU. Can I confirm that?"
Select your species of interest and your sequenced region. A red bracket means they cannot be told apart by that amplicon and are therefore collapsing into one OTU.
RADx includes RADport, a built-in guide that walks you through running your selected taxa through the MetaScope pipeline. Access it from the main menu after selecting your species.
RADx requires R >= 4.1. Core dependencies include:
shiny,bslib,shinyjs,shinyWidgetsggplot2,ggtext,plotlydplyr,tidyr,tibble,stringrBiostrings,msaRADalign
All dependencies are installed automatically via pak::pak("RADSuite/RADexplorer").
If you use RADx in your work, please cite the RADSuite package suite (citation tbd).
MIT