🌵Welcome to Drylab

Welcome to the documentation for a collection of tools (API and CLI) that I build to aid with the pain of locally reproducing published papers in computational biology.

Why did I build this

Most of science is not reproducible.

Locally reproducing papers is incredibly costly in terms of time. Usually, it requires reverse engineering analyses. This leads to significant effort wasted on reproducing papers that are important for our projects as bio researchers, and to the forgoing of the reproduction of other papers that might be important, but not enough to warrant all this effort.

I believe this is wrong architecture. Reproducibility is not “nice-to-have”, it should be an infrastructural requirement. With the tools I built to assist my research, now called Drylab, I am addressing this by automating the most common effort wasters for scientists related to reproducibility. We want to make it possible for scientists to order a paper to reproduce itself, and then self audit - to vet the outputs generated by the reproduced version against the paper's claims.

These tools will have succeeded if they have saved you, a researcher, significant time revenge engineering papers to reproduce them.