This is where I share negative, inconclusive, and otherwise unpublishable research results from projects of any field or scale.
An entry may document a quick test or one unresolved result from years of work. Each one records the original question, the experiment or proposed test, the result, what I learned, and why that line of work stopped. Code and data are included when available so the work can be inspected, reused, or continued. Ideas with no experiment remain labelled as ideas, and inconclusive results remain inconclusive.
Negative and inconclusive results still change what we know. They belong in the scientific record.
| Entry | Stage | Outcome | Source data |
|---|---|---|---|
| Enzyme solvent-tolerance prediction | completed | practical predictor claim not supported; biological hypothesis unresolved | OrganiZymeDB |
| Fly-connectome topology | idea | broad version already covered by prior work and too large for the intended side project | none |
Stage records how far the work progressed: idea, designed, attempted, completed, or extended.
Outcome records what the evidence supports: untested, supported, not-supported, inconclusive, invalidated, or superseded.
These labels are kept separate because an unperformed idea has no empirical outcome, and a failed assay says nothing about the hypothesis. Each completed entry also includes the code, data provenance, limitations, and the next experiment that would materially change the conclusion.
Corrections, reproductions, and extensions are welcome. See CONTRIBUTING.md.
The code, original text, and third-party data use different licenses. See LICENSES.md before reuse.