Venn diagram tool

You end up with lists from RNA-seq, proteomics, ChIP, CRISPR screens, or just two spreadsheets someone emailed you. The question is always some variant of: what’s shared, what’s unique, and how big is each bit?

We use this page when we want a quick visual answer without firing up R or fighting with BioVenn/Intervene for a simple three-way comparison. Typical jobs: genes up in treatment A vs B vs control, proteins in two pulldowns, or motif hits from different scanners. It is not a statistical test — it is counting exact string matches between lists you provide.

If you need fuzzy matching (same gene, different symbol) or ID conversion (Ensembl vs HGNC), clean the lists first in Excel or your favourite annotation tool, then paste here.

Each line in a text box is one item. We trim whitespace, drop blank lines, and collapse duplicates within a list. Two entries count as the same only if the trimmed text is identical — TP53 and tp53 are different; BRCA1 and BRCA1 (with a trailing space) are the same.

For every unique string across all lists, we work out which lists contain it and group strings into regions — e.g. “in list A and B but not C”. The Region breakdown table is the source of truth: region label, count, and every string in that region.

Diagram type by list count

• 2 or 3 lists — classic circle Venn diagrams with region counts (and optional example labels).
• 4 lists — four rotated ellipses (proper four-set Venn construction). Counts on the diagram are approximate for layout; use the table for exact membership.
• 5 to 8 lists — UpSet-style plot: one column per list, dots and lines show which lists share each overlap; counts at the right. The plot may show the top regions only if there are many; everything is still in the table.

1. Choose how many lists for the Venn diagram from the dropdown (2–8). Use + Add list / − Remove list to match — removed rows are stashed and come back if you add again.
2. Name each list (optional but helps in the legend) and paste one item per line.
3. Click Draw diagram. Check the status line for unique counts per list.
4. Scroll the Region breakdown table for exact overlaps. Use Export regions CSV for Excel or downstream analysis.
5. Tick Hide example strings on the 2–3 list circle diagrams if the plot is cluttered (counts and list names stay visible). For four ellipses this hides sample text in the centre overlap only; for five or more the UpSet plot already shows dots and counts only.

Tip: copy a single column from Excel and paste straight into a list box — no CSV upload needed.

• Diagram — quick picture of overlap sizes. For 2–3 sets you may see a few example gene names in each lobe unless hidden.
• Region breakdown table — every disjoint overlap region with a plain-English label (e.g. A ∩ B — not C), the count, and all member strings (newline-separated in the cell).
• Export regions CSV — same data as the table, UTF-8 with BOM for Excel. One row per region; the strings column can contain line breaks inside the cell.

The note under the tool repeats the main caveat: four-list ellipse counts are approximate on the figure; five-plus UpSet shows a subset of rows on the graphic if there are lots of regions. Trust the table for publishing numbers.

• Exact text only — no alias mapping, no case folding, no “within 1 kb” logic.
• Not weighted — a gene that is rank #1 and rank #5000 in two DE tables still counts as one overlap if the symbol matches.
• Large lists — thousands of lines work, but the table gets long; export CSV and filter in a spreadsheet if needed.
• Four-set diagram — pretty but approximate for counts on the SVG; always check the table.

• TF binding scanner — motif hits on a promoter (single sequence).
• Cross-species TF scanner — conserved motifs across orthologous promoters.
• Three-axis scatter plot — plot numeric columns from a CSV (e.g. PC scores) in 3D.
• Promoter Grabber — fetch promoter sequence before motif or overlap work on genomic loci.

Lists are parsed and compared entirely in your browser. Nothing is uploaded to our server for this tool. Do not paste identifiable patient data or confidential gene lists unless your policy allows. Site-wide cookies, analytics, and ads are described in the privacy policy.