Integrate two omics layers by correlation. Drop two feature tables (e.g. microbial features and metabolite features) — CorrOmics correlates every feature in one against every feature in the other across the samples they share, then hands you the significant pairs and a Cytoscape-ready network. Everything runs in your browser; the tables never leave your machine.
| Feature A | Feature B | r | |r| | n |
|---|
Each table is read as feature ID in the first column and one
numeric sample per remaining column; samples are paired by
matching column headers between the two tables. Correlations use only the
samples present and non-missing in both features. Spearman is Pearson on
ranks. No multiple-testing correction is applied — the |r| and sample-count
thresholds are the only filters, so treat surviving pairs as hypotheses.
The edge export is a plain CSV (source, target, correlation, abs_correlation, n, method)
that imports directly as a network in Cytoscape.