= 12 nM, biochemical, pH 7.4
Compare two measurements
Choose two records and the server checks whether they can be compared: identity, endpoint, value, assay, conditions, molecular form, dependence, direction, evidence class and sources. When a rule fails, the result names it and what would be needed.
Choose two records
Rehearse the rules with invented examples
Strictly matched metadataAll required identity, endpoint, condition, molecule and provenance fields match.Endpoint mismatchA Ki record and an IC50 record must remain distinct.Condition mismatchThe pH differs even though the other fixture fields match.Construct mismatchCanonical full-length and truncated constructs cannot be ranked as one target state.
These examples use invented metadata to show how the rules decide. They are not evidence and carry no scientific values or citations.
These can be compared
Strictly matched metadata: All required identity, endpoint, condition, molecule and provenance fields match.
= 18 nM, biochemical, pH 7.4
What this means
The two records match on every rule, so their values can be read side by side.
How the rules work
- Compare only exact, versioned protein identities; names are not identifiers.
- Rank quantitative measurements only within one assay or an independently reviewed matched-protocol group.
- Keep Ki, Kd, IC50, EC50, Km, kcat and kcat/Km as separate endpoint types.
- Preserve source relation operators and raw values; a bound is not a point estimate.
- Keep experimental constructs, variants, tested molecular forms and conditions explicit.
- Display a refusal with actionable reasons whenever strict comparison requirements are not met.