Methods

Transparent set arithmetic and explicit identifier handling.

VennScope compares observed set membership directly. Biological identifier conversion is a separate, explicit workflow so that a declared label is never silently treated as biological equivalence.

Input normalisation

Blank entries are removed and duplicate identifiers within a set are deduplicated. Matching is case-sensitive by default; users may explicitly choose case-insensitive comparison. VennScope does not silently uppercase biological identifiers.

Literal set comparison

For standard analysis, the comparison key is the identifier string supplied by the user after the selected matching rule is applied. Declaring an identifier format records metadata and enables plausibility checks; it does not itself convert an identifier or prove biological identity.

Gene, transcript and protein entities

Gene symbols, gene identifiers, transcript identifiers and protein identifiers represent related but distinct biological entities. A gene may correspond to multiple transcripts or protein products. Therefore a conversion from a gene-level identifier to a protein-level identifier can legitimately produce one-to-many results and can increase the number of identifiers in a derived dataset.

Identifier mapping and derived datasets

Identifier mapping is a separate, explicit action. Supported online routes use the named UniProt ID Mapping service; local/offline mapping uses a mapping table supplied by the user and does not require a network request. VennScope records the route and available provenance, displays mapped, one-to-many and unmapped records, preserves the original dataset, and creates a separate derived dataset only when the user requests it.

Mapped identifiers should be reviewed before analysis. A mapping result is not automatically equivalent to a one-to-one biological correspondence, particularly when moving between gene, transcript and protein levels.

Exact intersections

For sets A, B and C, an exact A ∩ B region contains identifiers present in A and B but not C. Exact regions correspond to complete membership signatures.

Inclusive intersections

Inclusive A ∩ B contains identifiers shared by A and B even when those identifiers also occur in additional displayed sets. The exact/inclusive selector changes the reported intersection definition, not the underlying parsed lists.

Venn and UpSet representation

Venn diagrams are recommended for two or three displayed sets. For larger comparisons, VennScope uses an UpSet representation because the number of possible intersections grows rapidly. Diagram geometry is schematic unless explicitly stated otherwise; numerical counts are derived from set membership rather than visual area.

Similarity metrics

Pairwise similarity includes shared-count, Jaccard, Dice and overlap coefficients. These coefficients describe different aspects of set similarity and should be interpreted with the set sizes and scientific context.

Overlap statistics

Probability-based overlap testing requires a scientifically appropriate background universe. VennScope reports a one-sided hypergeometric enrichment probability, a two-sided Fisher exact-test p-value, an odds ratio and a confidence interval for pairwise comparisons. A background universe should represent elements that could realistically have been observed in the experiment.

Plain identifier lists do not contain replicate quantitative measurements, so a t-test is not an appropriate test for ordinary set membership. Quantitative abundance data require a different data model and analysis.

Multiple testing

When multiple pairwise Fisher tests are displayed, VennScope reports Benjamini–Hochberg-adjusted q-values for that displayed family of tests. Statistical significance is not equivalent to biological importance, effect size or reproducibility.

Publication figures

The figure editor changes presentation properties such as typography, labels, margins, palette and output dimensions; it does not change the underlying set membership. Exported packages can include figure files, tables and provenance information according to the selected options.

Reproducibility

For consequential analyses, retain the original input lists, matching setting, declared identifier metadata, mapping CSV/provenance where mapping was used, background-universe definition for statistics, and the VennScope project/export package. Important results should be independently checked before publication.