PCA — Sample Clustering

Header: PCA: Sample Clustering

The PCA (Principal Component Analysis) tab provides an unsupervised overview of how your samples cluster relative to one another, helping you assess data quality and identify batch effects or outliers.

PCA Score Plot

The main scatter plot displays each sample as a colored dot, positioned according to its principal component scores. Samples that are more similar in their overall gene expression profile will appear closer together. The axes show the percentage of variance explained by each component (e.g., "PC1 (54.1%)").

The legend on the right side of the plot identifies sample groups by color.

You can change which principal components are plotted using the PCA Axes controls in the sidebar (X axis and Y axis dropdowns).

Scree Plot

Below the PCA Score Plot, a Scree Plot displays the variance explained by each principal component as blue bars, with a cumulative variance line (red) overlaid. This helps you assess how many components are needed to capture the majority of variance in the data.

• X-axis: Principal components (PC1 through PC10)
• Left Y-axis: Individual variance explained (%)
• Right Y-axis: Cumulative variance (%)

PC Scores Table

An expandable data table below the Scree Plot shows the numerical PC scores for every sample across all computed components (PC1–PC10), along with each sample's group assignment. This table is useful for exporting or verifying exact values.