ideasIdea

A registry for preclinical experiments that did not work

Most lab experiments that fail are never written up, so other labs repeat them. A simple, structured registry with a citable record for each failed experiment would stop the waste.

Negative preclinical findings (a drug that did not work in a model, a target that did not validate, a biomarker that did not predict) have no home. A registry with minimal structured fields (target, model, intervention, readout, result, methods link, raw data), each with a DOI, and integrated into the major databases, would make failures findable. Funders would require deposit for supported projects.

Hypothesis

Within three years, at least 20% of active oncology labs in participating institutions deposit at least one record per year, and surveyed researchers report avoiding at least one planned experiment because of a registry entry.

Rationale

The gene-target level bias is enormous: a handful of targets attract most papers, partly because negative results on them are invisible. Registries with low friction (such as bioRxiv for preprints) achieved rapid adoption once norms shifted.

What would test it

Launch with three funders requiring deposit; measure deposits, views and self-reported behaviour change after two years.

Maturity

speculative

Who has to act

research

Cost to try

Small (under $1M)

Years to first evidence

Bottlenecks it attacks

Failures are hidden · Negative trials, failed drugs and abandoned programmes are rarely published, so the same mistakes are repeated.
Preclinical results do not reproduce · Fewer than half of landmark cancer biology findings reproduce when someone else tries.

Connected

10top

Pages like this

not linked directly; found by shared links

A registry for preclinical experiments that did not work

Connected

Pages like this

ideas

collections

bottlenecks