[R] predictive logistic model cell-biology, non-dichotomous data

[andreasss]

Hi everyone,

I would like to fit a predictive model to my data in order to compare
absorbance readings to a toxin standard. This data was obtained by exposing
red blood cells to different toxin concentrations, which lead to the lysis
of the red blood cells, increasing the absorbance (hemoglobin is freed). The
data has a sigmoid shape (see below), so I thought about fitting a logistic
model to the data so that I will be able to determine the toxin equivalent
of new absorbance readings.
http://r.789695.n4.nabble.com/file/n3595812/Unbenannt.jpg 
 
The data points for this curve are:
http://r.789695.n4.nabble.com/file/n3595812/qweqwe.jpg 
I must admit that I am totally lost. I have done a fair bit of reading on
logistic regression, but most seem to focus on binary outcomes or
multinomial analysis. Do I have to somehow assign 'pass' or 'fail' to this
data, maybe 0 and 100% lysis? Or is the logistic model not suitable for what
I am planning. All I want to do is to fit a predictive model to this data
and to graphically represent the 'best fit'. Any help will be greatly
appreciated.

Thanks in advance,

Andreas


--
View this message in context: http://r.789695.n4.nabble.com/predictive-logistic-model-cell-biology-non-dichotomous-data-tp3595812p3595812.html
Sent from the R help mailing list archive at Nabble.com.