TimelyScholarFree Tools › PCA Plot Generator
Run principal component analysis on your data and plot the samples on the first two principal components, coloured by group, with the percentage of variance each component explains.
FreeNo account neededYour data stays in your browser
| PC1 | 89.1%Cumulative 89.1% |
|---|---|
| PC2 | 4.1%Cumulative 93.2% |
| PC3 | 3.3%Cumulative 96.5% |
| PC4 | 2.2%Cumulative 98.7% |
| PC5 | 1.3%Cumulative 100.0% |
Scaling gives every variable equal weight; without it, high-variance variables dominate. Axis signs are arbitrary. Results match scikit-learn's PCA.
Samples in columns (expression tables) or rows.
Covariance (centre only) or correlation (centre and scale).
The plot and the PC scores.
Principal component analysis finds new axes, the principal components, that capture as much of the variation in the data as possible. Plotting samples on the first two components shows which samples are similar overall.
In expression studies, PCA is often used to check whether samples group by condition and to spot outliers or batch effects.
Scale (use the correlation matrix) when variables are measured in different units or on very different scales; otherwise variables with large variances dominate the first components.