Lab Note · updated 2026-06-05

Percent Replicating

The reproducibility metric that catches a broken assay before it reaches a biologist.

TopicsCorrect · Readout · Cell Painting · copairs

Percent replicating asks a blunt question: do replicate perturbations look more like each other than like random pairs? If not, stop — something upstream broke.

It only means something after Illumination Correction Before You Trust a Profile and Batch Correction Without Erasing Biology.

Related notes

From Masks to Morphology: Features & Self-Supervised Embeddings
Once cells are segmented, you turn masks into numbers — either hand-engineered morphology features or learned embeddings. Each choice trades interpretability against the signal it can capture.
Why Microscopy Needs Verified Outputs
A measurement you cannot reproduce is not a measurement. Why verification is a first-class pipeline stage, not an afterthought.
Illumination Correction Before You Trust a Profile
Uneven illumination silently biases every downstream feature. Correct it first, or your batch effects are really optics.
Batch Correction Without Erasing Biology
Sphering and Harmony remove plate effects — push too hard and you remove the signal too.

Local graph

Related notes

From Masks to Morphology: Features & Self-Supervised Embeddings

Why Microscopy Needs Verified Outputs

Illumination Correction Before You Trust a Profile

Batch Correction Without Erasing Biology