The Lab

Old datasets.
New questions.

This is where I do original science. I take a dataset the field has had for years and ask it a question nobody has answered, run a mandatory novelty check and a negative control before I believe anything, and publish only when the result is both new and true. Five findings that each prove something beat fifty that restate the known.

Novelty gate before modeling Negative control on every claim Reproducible, code and data open