Microbiome & multi-omics / Procrustes analysis
Your data stays in your browser: files are read and processed on your device, never uploaded to any server, and cleared when you close the page. Free, no sign-in.
Procrustes analysis superimposes two ordinations (for example a PCA of species composition and a PCA of environmental variables) by translation, rotation, reflection and uniform scaling, and measures their disagreement with the residual sum of squares m²: the smaller m² (the larger r = √(1 − m²)), the more similar the sample patterns.
Each sample has two points on the plot: its position in data set 1 and in the rotated data set 2; shorter lines mean better agreement. The residual table lists the least concordant samples.
The PROTEST permutation test shuffles the sample order of data set 2; P shows whether the observed agreement is stronger than for random pairings. A fixed seed makes it reproducible.
Samples are matched by name and only those present in both data sets are used. If you already have ordination coordinates (e.g. PCoA or NMDS scores), choose “Already ordination coordinates — use as is”.
Gower J. C. (1975) Psychometrika 40:33–51; Jackson D. A. (1995) Écoscience 2:297–303 (PROTEST); Peres-Neto & Jackson (2001) Oecologia 129:169–178. m² and rotated coordinates match scipy.spatial.procrustes; PCA matches scikit-learn.
Up to 5,000 samples.
Send us your data and research question and you will receive a written plan within 1 working day: analysis steps, parameter rationale, deliverables and timeline. Quoted per project.