How to use (3 steps)
- Choose an example, or paste an OTU/ASV table (or import a CSV/TSV file).
- Select a metric (Jaccard / Bray–Curtis) and preprocessing (relative abundance, etc.).
- Review the distance matrix, heatmap, and PCoA (export CSV/PNG if needed).
Beta diversity is useful for exploration and visualization, but statistical conclusions require additional analyses (tests or modeling).
Inputs
—
Metric & preprocessing
If sample depths differ, try “Relative abundance”.
Sample groups (optional: color points in PCoA)
Two columns: sample, group (header is optional).
Results
This tool is for exploration and learning. It does not claim statistical significance.
—
Share & export
The share URL restores settings only (input data is not included). To save inputs too, use JSON export.
Distance matrix
Heatmap
PCoA (2D)
Jaccard vs Bray–Curtis
- Jaccard compares presence/absence only.
- Bray–Curtis also uses abundances (counts or relative abundance).
Bray–Curtis can be affected by library size (sample depth). If needed, compare both counts and relative abundance.
Equations (reference)
- Jaccard distance:
d = 1 - |A∩B| / |A∪B| - Bray–Curtis dissimilarity:
d = Σ|xᵢ - yᵢ| / Σ(xᵢ + yᵢ)
Here, A and B are sets of observed features, and xᵢ/yᵢ are feature values (relative abundance or counts). Axis directions are arbitrary, so flipping the PCoA plot does not change the meaning.
FAQ
What is beta diversity?
A way to describe how different samples are from each other as a distance (dissimilarity). Smaller means more similar; larger means more different.
What is the difference between Jaccard and Bray–Curtis?
Jaccard compares presence/absence only. Bray–Curtis also uses abundances (counts or relative abundance).
What is PCoA?
A method that places samples in 2D based on the distance matrix (principal coordinates analysis). Axis direction is arbitrary, so flipping does not change the meaning.
Can I claim statistical significance from this result alone?
No. Beta diversity is useful for exploration and visualization, but statistical testing requires additional analyses.
What data should I enter first for beta diversity?
Start with comparable sample count tables using the same taxa or categories. Verify zeros, totals, and sample names before comparing distance, similarity, or ordination outputs.
Related tools
- Check within-sample diversity (alpha diversity) first → Diversity index calculator (Shannon / Simpson)
- Build intuition for population size and drift → Genetic drift simulator (Wright–Fisher)
- Diversity index calculator (Shannon & Simpson) | CalcBEEstimate within-sample diversity first, then compare samples with beta-diversity distance metrics.
- NGS coverage calculator | reads, target size, depth, ≥k× | CalcBEPlan sequencing depth and usable coverage before you interpret between-sample community differences.
- Growth curve calculator | doubling time, growth rate, lag model | CalcBEUse a complementary biology workflow when you need time-course interpretation rather than community distance.
Comments
If you find an issue or have a suggestion, please leave a comment (no anonymous form).