Upset Plots

What the source does

• FilesystemNone observedNo reads or writes in the source
• NetworkNone observedNo outbound calls in the source
• ShellNone observedNo commands run in the source
• DependenciesPinnedNothing floats
• SecretsNone foundNo keys in source

Five audit layers

Gates applied: no behavioural pass.

Audit 0.4.1 · source sha c8c1fc3a2b13 · 19 Oct 2026Report an issue or request a re-scan

• PublisherGPTomics GitHub organizationOwner of the repository this skill is read from
• Source folderGPTomics/bioSkills/data-visualization/upset-plotsThe exact folder we audited
• LicenceMITAs the repository states it
• Commit pinned for this verdictc8c1fc3a2b13A new audit runs when the source changes
• Repository statusArchivedLast push 65 days ago
• Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

Its description: Build UpSet plots to visualize set intersections beyond 4 sets (where Venn fails) using ComplexUpset (modern, ggplot2-grammar) or the unmaintained UpSetR, with explicit cardinality vs degree sorting, attribute panels,…

---
name: bio-data-visualization-upset-plots
description: Build UpSet plots to visualize set intersections beyond 4 sets (where Venn fails) using ComplexUpset (modern, ggplot2-grammar) or the unmaintained UpSetR, with explicit cardinality vs degree sorting, attribute panels, and query highlighting. Use when comparing overlap across many gene sets, peak sets, variant lists, or any set membership matrix where Venn diagrams become illegible.
tool_type: mixed
primary_tool: ComplexUpset
---

## Version Compatibility

Reference examples tested with: ComplexUpset 1.3+ (R, Krassowski), UpSetR 1.4.0 (last 2019 release; effectively unmaintained), upsetplot 0.9+ (Python).

Before using code patterns, verify installed versions match. If versions differ:
- R: `packageVersion('<pkg>')` then `?function_name`
- Python: `pip show <package>` then `help(module.function)`

If code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API rather than retrying.

# UpSet Plots

**"Show set intersections for 4+ sets"** -> Replace Venn diagrams (which become illegible past 4 sets) with UpSet (Lex 2014 *IEEE TVCG* 20:1983). The display: a matrix of dots indicating which sets participate in each intersection, with a vertical bar above each column showing intersection size and horizontal bars on the left showing per-set total size. Sort by intersection size (cardinality) for "biggest overlaps first" or by degree (number of sets) for grouped layout.

- R: `ComplexUpset::upset` (Krassowski; ggplot2-native, **recommended**), `UpSetR::upset` (Conway 2017; legacy, unmaintained)
- Python: `upsetplot.UpSet`

## The Single Most Important Modern Insight -- UpSetR Is Effectively Unmaintained

UpSetR (Conway, Lex, Gehlenborg 2017 *Bioinformatics* 33:2938) is the original R implementation but has had no CRAN release since v1.4.0 (2019). ComplexUpset (Krassowski; CRAN active through 2025-07) is the actively maintained ggplot2-grammar replacement. For new work in 2026, **prefer ComplexUpset**. Caveat: ggplot2 4.0 (mid-2025) broke ComplexUpset's `upset()` function (issue #213); pin to compatible versions until patched.

The Lex 2014 paper and underlying UpSet visualization concept are not affected — the visualization is the same; the difference is which R package implements it best in the current ecosystem.

## ComplexUpset (Modern Default)

**Goal:** Render a set-intersection plot with cardinality-sorted bars, optional metadata stacks (e.g., percent of intersection significant), and pre-specified queries highlighting biologically relevant intersections.

**Approach:** Convert set memberships to a long-format data frame with one row per element and binary columns per set; pass to `upset()` with `intersections='all'` or pre-specified subset; use `ComplexUpset::upset_query` to highlight intersections.

```r
library(ComplexUpset)
library(ggplot2)

# Convert from list of sets to long format
sets <- list(SetA = c('Gene1','Gene2','Gene3','Gene4'),
             SetB = c('Gene2','Gene3','Gene5','Gene6'),
             SetC = c('Gene1','Gene3','Gene6','Gene7'),
             SetD = c('Gene3','Gene4','Gene7','Gene8'))

# Long-format binary membership matrix
all_elements <- unique(unlist(sets))
df <- data.frame(element = all_elements)
for (s in names(sets)) df[[s]] <- df$element %in% sets[[s]]

# UpSet
upset(df,
      intersect = names(sets),                          # which columns are sets
      n_intersections = 20,                             # show top 20 intersections
      sort_intersections = 'descending',                # by cardinality
      sort_intersections_by = 'cardinality',            # 'cardinality' OR 'degree'
      base_annotations = list(
          'Intersection size' = intersection_size(
              counts = TRUE,
              text = list(size = 3))),
      themes = upset_modify_themes(
          list('Intersection size' = theme(panel.grid = element_blank()))))
```

## Sorting -- Cardinality vs Degree

**Cardinality sort** (default): intersections ordered by size (largest first). Reveals "the biggest overlap is A∩B."

**Degree sort**: intersections grouped by *number of sets they include* (1-set intersections, then 2-set, then 3-set, etc.). Reveals "how distributed are the overlaps across set counts?"

Choose based on the scientific question. Cardinality is the default for "find the biggest overlap"; degree is appropriate when comparing across "exclusive to 1 set" vs "shared by all."

## Pre-Specified Queries / Highlighting

```r
upset(df,
      intersect = names(sets),
      queries = list(
          upset_query(intersect = c('SetA', 'SetB'),
                       color = '#D55E00', fill = '#D55E00',
                       only_components = c('intersections_matrix', 'Intersection size')),
          upset_query(intersect = c('SetA', 'SetC', 'SetD'),
                       color = '#0072B2', fill = '#0072B2',
                       only_components = c('intersections_matrix', 'Intersection size'))))
```

## Attribute Panels (ComplexUpset Strength)

Unlike UpSetR's "boxplot.summary," ComplexUpset supports arbitrary ggplot annotations stacked above the intersection bars:

```r
upset(df,
      intersect = names(sets),
      annotations = list(
          'log2 FC' = ggplot(mapping = aes(x = intersection, y = log2FC)) +
                       geom_boxplot() + theme_classic(),
          'Significant fraction' = ggplot(mapping = aes(x = intersection, fill = significant)) +
                                    geom_bar(position = 'fill') +
                                    scale_fill_manual(values = c('TRUE' = '#D55E00', 'FALSE' = 'grey80')) +
                                    theme_classic()))
```

## upsetplot (Python)

```python
from upsetplot import from_contents, UpSet
import matplotlib.pyplot as plt

sets = {'SetA': ['Gene1','Gene2','Gene3','Gene4'],
        'SetB': ['Gene2','Gene3','Gene5','Gene6'],
        'SetC': ['Gene1','Gene3','Gene6','Gene7']

Excerpt, first 6,000 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

pip install upsetplot

git clone https://github.com/GPTomics/bioSkills
git -C bioSkills checkout c8c1fc3a2b13

Works in

1. 2026-10-19
   ✓SAFE B 89 / 100

   first audit · source c8c1fc3a2b13

Questions about Upset Plots

• github.com/GPTomics/bioSkills/tree/main/data-visualization/upset — the audited source at commit c8c1fc3a2b13 and every finding aboveObserved · 19 Oct 2026

Spotted something wrong? Report a correction · Methodology