Fig. 1

Workflow of omicplotR. The data input requires samples as columns, features by rows, whereas the metadata input descriptors as columns and samples by row. Count tables can be filtered to remove samples or features with low counts. After filtering, zero-imputation and a log-ratio transform is applied to the counts. A principal component analysis (PCA) biplot is typically the first exploratory visualization used. Several other plots are available to visualize differences between samples, features, and experimental conditions. Visualizing which features and samples have been removed by filtering is also possible. Plots are stylized representations of plots that can be generated by omicplotR