---
title: Functional Annotation
slug: functional-annotation
createdAt: 2026-04-27T07:51:15.046Z
---

Functional annotation shows how each dataset contributes to the edge weights in the graph. This gives an understanding of how different datasets and sample categories contribute to the connections in the network.

The Group By dropdown controls how the datasets are grouped. By default, the view is grouped by Data Source, where each row represents one dataset. Datasets can also be grouped by sample categories, such as biopsy sites, cell types, or diseases. This is helpful to understand how different datasets and/or sample categories contribute to the graph processes.

For each dataset listed, detailed information is given, such as the number of samples included during network construction, variable category inclusions, and a summary of the data. This provides a deeper understanding of the data behind the connections.

**How to read the functional annotation**

Each row represents one dataset, and each vertical bar within a row represents that dataset's contribution to one edge in the graph. When you click an edge in the graph view, its corresponding bar is highlighted.

Bar positions are based on z-scores calculated separately for each dataset, using the dataset's correlation coefficients across all edges in the filtered network. Bars further to the right show a stronger positive contribution to the edge weight.

- **p value:** The p value to the right of each row shows whether the dataset's z-score is above the median of the other supporting datasets.
- **Breakpoint:** The dotted line shows the dataset's 99.9th percentile z-score. Bars to the right of it are the top 0.1% most supported edges for that dataset.

**Explore the chart**

- Hover over a highlighted bar to see the z-score for the selected edge.
- Use the slider below the x-axis to narrow the z-score range shown.