Using of R in visualbasic or ...

[Uranne]

Hello
I'm in internship and I have to make some analysis on microbial kinetics. I have to automise the treatment with vb.net but to work on kinetic, i have to use R.

On R, I have to found the inflexion point of the exponential, in order to found, for exemple, a retard caused by a product.
To do it, I use the package named "drc", the function drm for a L.3 regression, but I when I declare drc, it doesn't work, this is my code in vb :

REngine.SetDllDirectory("C:\Program Files\R\R-2.15.2\bin\i386") 
        Dim engine = REngine.CreateInstance("RDotNet") 
        'engine.EagerEvaluate("C:\Program Files\R\R-2.15.2\library\""" & ")") 

        engine.EagerEvaluate("library(plotrix)") 
        engine.EagerEvaluate("library(abind)") 
        engine.EagerEvaluate("library(magic)") 
        engine.EagerEvaluate("library(gtools)") 
        engine.EagerEvaluate("library(MASS)") 
        engine.EagerEvaluate("library(nnet)") 
        engine.EagerEvaluate("library(car)") 
        engine.EagerEvaluate("library(alr3)") 
        engine.EagerEvaluate("library(lattice)") 
        engine.EagerEvaluate("library(nlme)") 
        engine.EagerEvaluate("library(stats4)") 
        engine.EagerEvaluate("library(drc)")

The firsts work, but MASS doesn't work, and drc need all those packages to by used.
It works on R STUDIO but not by vb ...

If you can't help me, do you have a better method to found this inflexion point ?

[dpryan]

I've never tried to interface R and visual basic (just python and C), so I can't help there.

An alternative approach to the drc package would be to just use lm. After all, kinetics are exponential processes which become linear when one takes the natural log. BTW, since this isn't really a sequencing related question, you might also look into the Biostars forum for similar questions in the future.

[Uranne]

Sorry if I haven't chose the right forum ... I didn't know Biostars ...

About using lm, I don't understand how to use it. Have I to take the log of my datas before use it ?
But it isn't a question to ask here, thank you for your help.

[dpryan]

Sorry if I haven't chose the right forum ... I didn't know Biostars ...

About using lm, I don't understand how to use it. Have I to take the log of my datas before use it ?
But it isn't a question to ask here, thank you for your help.

Yeah, if y=a*exp(-x/b) then ln(y) = ln(a) - x/b, which is a simple linear regression (ln is the natural log, to avoid confusion).

[Uranne]

mmmh...
I understand the logic and I did it in R to study it. But where in this case do I found the abscissa of my inflexion point ? The summary gives the intercept but it is not the good value.