Differential analysis & tests / Correlation 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.
Pearson’s r measures linear association and is sensitive to outliers; Spearman’s and Kendall’s coefficients use ranks, measure monotonic association and are more robust to outliers and non-normality.
With one table, all pairs of the ticked variables are correlated; with a second table, variables of the two tables are correlated with each other (e.g. gene expression with metabolites, taxa with environmental factors).
Each pair uses the samples where both values are present (pairwise deletion). P values are adjusted across all pairs; Pearson’s r also gets a Fisher-z confidence interval.
Heatmap colour is the coefficient (red positive, blue negative); non-significant cells can be left blank. Correlation is not causation, and coefficients are unstable in small samples.
Pearson and Spearman coefficients and P values match scipy.stats.pearsonr / spearmanr; Kendall tau-b matches scipy.stats.kendalltau; Pearson CIs use the Fisher z transformation; adjustment matches R p.adjust.
Up to 50,000 rows; many variable pairs (hundreds of variables) take longer.
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.