Best practice building non-linear brms models?

Interfaces brms
5

Finally,

I would like to predict and show how my parameters tg, k, and Gm vary with factor values:

lntg ~ (cacl + dose + pH + temperature + protein)^2 + I(cacl^2) + I(dose^2) + I(pH^2) + I(temperature^2) + I(protein^2)

lnGm ~ (cacl + dose + pH + temperature + protein)^2 + I(cacl^2) + I(dose^2) + I(pH^2) + I(temperature^2) + I(protein^2)

lnk ~ (cacl + dose + pH + temperature + protein)^2 + I(cacl^2) + I(dose^2) + I(pH^2) + I(temperature^2) + I(protein^2)

I know how to do this by extracting the draws and combining by ‘hand’, but was wondering if there is an easier way to do this?

e.g something like:

parameter = predict(fit, newdata = newdata, nl_parameter_to_predict = ‘lnk’) …

Thanks

/Jannik