PMID- 22961887
OWN - NLM
STAT- MEDLINE
DCOM- 20130509
LR  - 20121204
IS  - 1520-6033 (Electronic)
IS  - 1520-6033 (Linking)
VI  - 28
IP  - 6
DP  - 2012 Nov-Dec
TI  - Statistical vs. stochastic experimental design: an experimental comparison on the 
      example of protein refolding.
PG  - 1499-506
LID - 10.1002/btpr.1635 [doi]
AB  - Optimization of experimental problems is a challenging task in both engineering 
      and science. In principle, two different design of experiments (DOE) strategies 
      exist: statistical and stochastic methods. Both aim to efficiently and precisely 
      identify optimal solutions inside the problem-specific search space. Here, we 
      evaluate and compare both strategies on the same experimental problem, the 
      optimization of the refolding conditions of the lipase from Thermomyces 
      lanuginosus with 26 variables under study. Protein refolding is one of the main 
      bottlenecks in the process development for recombinant proteins. Despite 
      intensive effort, the prediction of refolding from sequence information alone is 
      still not applicable today. Instead, suitable refolding conditions are typically 
      derived empirically in large screening experiments. Thus, protein refolding 
      should constitute a good performance test for DOE strategies. We compared an 
      iterative stochastic optimization applying a genetic algorithm and a standard 
      statistical design consisting of a D-optimal screening step followed by an 
      optimization via response surface methodology. Our results revealed that only the 
      stochastic optimization was able to identify optimal refolding conditions (~1.400 
      U g(-1) refolded activity), which were 3.4-fold higher than the standard. 
      Additionally, the stochastic optimization proved quite robust, as three 
      independent optimizations performed similar. In contrast, the statistical DOE 
      resulted in a suboptimal solution and failed to identify comparable activities. 
      Interactions between process variables proved to be pivotal for this 
      optimization. Hence, the linear screening model was not able to identify the most 
      important process variables correctly. Thereby, this study highlighted the limits 
      of the classic two-step statistical DOE.
CI  - Copyright (c) 2012 American Institute of Chemical Engineers (AIChE).
FAU - Anselment, Bernd
AU  - Anselment B
AD  - Institute of Biochemical Engineering, Faculty of Mechanical Engineering, 
      Technische Universitat Munchen, Boltzmannstrasse 15, Garching, Germany.
FAU - Schoemig, Veronika
AU  - Schoemig V
FAU - Kesten, Christopher
AU  - Kesten C
FAU - Weuster-Botz, Dirk
AU  - Weuster-Botz D
LA  - eng
PT  - Journal Article
PT  - Research Support, Non-U.S. Gov't
DEP - 20121018
PL  - United States
TA  - Biotechnol Prog
JT  - Biotechnology progress
JID - 8506292
RN  - 0 (Fungal Proteins)
RN  - 0 (Proteins)
RN  - EC 3.1.1.3 (Lipase)
SB  - IM
MH  - Ascomycota/chemistry/enzymology
MH  - Biotechnology
MH  - Fungal Proteins/chemistry/metabolism
MH  - Lipase/chemistry/metabolism
MH  - Models, Molecular
MH  - *Models, Statistical
MH  - *Protein Refolding
MH  - Proteins/*chemistry/metabolism
MH  - Research Design
MH  - *Stochastic Processes
EDAT- 2012/09/11 06:00
MHDA- 2013/05/10 06:00
CRDT- 2012/09/11 06:00
PHST- 2012/05/31 00:00 [received]
PHST- 2012/07/23 00:00 [revised]
PHST- 2012/09/11 06:00 [entrez]
PHST- 2012/09/11 06:00 [pubmed]
PHST- 2013/05/10 06:00 [medline]
AID - 10.1002/btpr.1635 [doi]
PST - ppublish
SO  - Biotechnol Prog. 2012 Nov-Dec;28(6):1499-506. doi: 10.1002/btpr.1635. Epub 2012 
      Oct 18.