Premium
Detection, Quantification, and Propagation of Uncertainty in High‐Throughput Experimentation by Monte Carlo Methods
Author(s) -
Osberghaus A.,
Baumann P.,
Hepbildikler S.,
Nath S.,
Haindl M.,
von Lieres E.,
Hubbuch J.
Publication year - 2012
Publication title -
chemical engineering and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.403
H-Index - 81
eISSN - 1521-4125
pISSN - 0930-7516
DOI - 10.1002/ceat.201100610
Subject(s) - monte carlo method , throughput , propagation of uncertainty , computer science , sampling (signal processing) , uncertainty quantification , in silico , biological system , algorithm , chemistry , machine learning , mathematics , statistics , detector , biology , telecommunications , biochemistry , wireless , gene
Since the efficiency and speed of computing has increased significantly in the last decades, in silico‐approaches, e.g., quasi‐experimental analyses based on mechanistic simulations combined with Monte Carlo (MC) methods, are on the rise for uncertainty analyses and estimation of uncertainty propagation. The power and convenience of these approaches for high‐throughput processes will be demonstrated with a case study including miniaturized screenings on robotic platforms: a binding study for lysozyme on the adsorbent SP Sepharose FF in 96‐well format. All relevant uncertainties during the experimental preparations and automated high‐throughput experimentation were identified, quantified, and then embedded in a simulation algorithm for the calculation of uncertainty propagation based on MC sampling. A proof‐of‐concept for this approach is then followed by the simulation‐based analysis of various case scenarios. The MC‐based approach can easily be transferred to uncertainty analyses in other high‐throughput processes.