Antibody Production Optimization Example

DESCRIPTION:

This is a response surface method experiment to find the treatment conditions that optimize the yield of monoclonal antibodies produced in mice innoculated with hybridoma cells. The response is proportional to the number of antibody molecules produced. The experimental design is a central composite design which has 2 experimental factors each at five levels. The design consists of 11 runs and includes 3 center points.

See Chapter 2 of Haaland (1989) for a more complete description of this experiment. See also the help file for abscrn.df for a description of the screening experiment that preceded this optimization experiment.

ARGUMENTS:

RadDos
an experimental factor giving the radiation dose in rads.
Prime1
an experimental factors giving the time in days between the initial injection of Pristane oil and the innoculation with antibody producing cells.
TtrVol
the response variable, antibody titer adjusted for volume. The measured value is proportional to the number of monoclonal antibody molecules produced. The response should be maximized.

REFERENCES:

Haaland, P. D. (1989). Experimental Design in Biotechnology. New York: Marcel Dekker, Chapter 2.

SOURCE:

Used by permission of Marcel Dekker, Inc.

SEE ALSO:

EXAMPLES:

# This design is already available in S-PLUS under  
# the name abrsm.df. These are the commands, that were 
# used to create it. 
abrsm.design <- rsm.design(2,factor.names =  
     list(RadDos=c(100,300),Prime1=c(7,21))) 
abrsm.ttrvol <- c(207,257,306,570,100,513,315, 
     154,630,528,609) 
abrsm.df <- cbind(abrsm.design,TtrVol=abrsm.ttrvol) 
# Sample analysis 
abrsm.rsm <- rsm.lm(abrsm.df) 
summary(abrsm.rsm) 
pareto(abrsm.rsm) 
contour(abrsm.rsm,levels=c(100,200,300,400,500, 
     550,590,610))  
surface(abrsm.rsm)  
optim(abrsm.rsm)