Why does the product say one thing and the joint p another?

The product is an e-value only when the tests are independent. A gene that scores high in one table tends to score high in the others, so the joint p, which ranks the product among every gene, is the safer reading on real tables.

Why does a test with no evidence lower the product?

A gene with no evidence in a table ties at the bottom with every other such gene and gets p = 1, which the calibrator turns into an e-value of 0.5.

Is anything uploaded?

No. The tables are ranked in the browser.

Analyze

Target validation: test "gene G drives phenotype P" against gene-level evidence tables

Test a hypothesis that a gene drives a phenotype, such as "RAC2 regulates IL-2 production", the way POPPER does. Each evidence table is one falsification test: an implication of the hypothesis that a gene-level score could contradict. The gene's p in each table is its rank among every gene the tables score, each p becomes an e-value, and the hypothesis passes when their product exceeds 1/α.

Open Target validation: test "gene G drives phenotype P" against gene-level evidence tables Open the Target validation sample

Does RAC2 regulate IL-2 production? Three falsification tests from GTEx, the GWAS Catalog and BioGRID, combined as e-values the way POPPER does

What it does

01One falsification test per evidence table: the gene's rank among every gene the tables score
02Vovk–Wang calibration of each p to an e-value and their product, as POPPER combines tests
03A joint p from the product's own rank among every gene, valid when the tables are correlated
04The test after which the evidence first crossed 1/α

What it runs on

Every gene's ranks are checked against pandas and every product against POPPER's own e-value calibrator, on all 18,706 genes of the bundled tables. The panel runs on @ports/workflow.

Everything runs client-side. Files are parsed in the browser and projects are saved on this machine, so an unpublished sequence is never uploaded.

Questions

Why does the product say one thing and the joint p another?: The product is an e-value only when the tests are independent. A gene that scores high in one table tends to score high in the others, so the joint p, which ranks the product among every gene, is the safer reading on real tables.
Why does a test with no evidence lower the product?: A gene with no evidence in a table ties at the bottom with every other such gene and gets p = 1, which the calibrator turns into an e-value of 0.5.
Is anything uploaded?: No. The tables are ranked in the browser.

Also searched as

target validation · POPPER hypothesis validation · falsification test gene phenotype · does this gene regulate · e-value hypothesis test · sequential falsification · gene prioritisation from GTEx GWAS BioGRID