Dose-response model with partial pooling

[jroon]

Apparently sigmoid curves are tricky for identifiability reasons.

See these links:
https://www.martinmodrak.cz/2018/05/14/identifying-non-identifiability/

Difficulties with logistic population growth model

Me and @stemangiola eventually ended up using a different reparametrization of the sigmoid for a very similar use case (there is a paper in final stages of preparation including that): f(x) = \frac{y_0 (1 + exp( \beta M))}{1 + exp(\beta (M - x ))} in this parametrization, y_0 = f(0) and \beta = \frac{4D}{A} is “slope” parameter. Those parameters are in our experience mostly well identified. Note that, we centered and scaled x before fitting the model, so y_0 is the value at the midpoint of d…

Dissection of prostate tumour, stroma and immune transcription reveals a key...

Struggling with all this myself!