Frederic Clerc

Library design using genetic algorithms for catalyst discovery and optimization

This study reports a detailed investigation of catalyst library design by genetic algorithm (GA). A methodology for assessing GA configurations is described. Operators, which promote the optimization speed while being robust to noise and outliers, are revealed through statistical studies. The genetic algorithms were implemented in GA platform software called OptiCat, which enables the construction of custom-made workflows using a tool box of operators. Two separate studies were carried out (i) on a virtual benchmark and (ii) on real surface response which is derived from HT screening. Additionally, we report a methodology to model a complex surface response by binning the search space in small zones that are then independently modeled by linear regression. In contrast to artificial neural networks, this approach allows one to obtain an explicit model in an analogical form that can be further used in Excel or entered in OptiCat to perform simulations. While speeding the implementation of a hybrid algorithm...

Development of an integrated informatics toolbox: HT kinetic and virtual screening.

We discuss thoroughly aspects and issues for the development of a bespoke, but generic, electronic infrastructure designed to cope with the dynamic in high-throughput experimentation and knowledge management, is applicable to large or contract research organizations. We present the first generation of an informatics platform developed for TOPCOMBI, a research project funded by the European Commission for Nanotechnology and Nanoscience. It is composed by an infrastructure and a collection of modules dealing with laboratory analytics, robotics, data handling and analytics, optimization, in-database processing and visualization, which are developed collegially by the partners of the Consortium. This best-of-breed informatics system enables the capture and the re-usage of processes and methodologies, i.e. process and data flows, using the workflow paradigm. Complex workflows designed by power users can be eventually used by either other domain experts or by novices through a web portal. Workflows can also be run interactively to allow visual analytics for instance, or automatically. We present two case studies dealing with the kinetic study of glycerol catalytic oxidation using parallel equipments, and a novel, fully integrated QSAR applied in heterogeneous catalysis, respectively.

Computational Methods in the Development of a Knowledge-Based System for the Prediction of Solid Catalyst Performance

The objective of this work is the construction of a correlation between characteristics of heterogeneous catalysts, encoded in a descriptor vector, and their experimentally measured performances in the propene oxidation reaction. In this paper the key issue in the modeling process, namely the selection of adequate input variables, is explored. Several data-driven feature selection strategies were applied in order to obtain an estimate of the differences in variance and information content of various attributes, furthermore to compare their relative importance. Quantitative property activity relationship techniques using probabilistic neural networks have been used for the creation of various semi-empirical models. Finally, a robust classification model, assigning selected attributes of solid compounds as input to an appropriate performance class in the model reaction was obtained. It has been evident that the mathematical support for the primary attributes set proposed by chemists can be highly desirable.

Optimisation Methodologies and Algorithms for Research on Catalysis Employing High-Throughput Methods: Comparison Using the Selox Benchmark

The Selox is a catalytic benchmark for the selective CO oxidation reaction in the presence of H(2), in the form of mathematical equations obtained via modelling of experimental results. The optimisation efficiencies of several Global Optimisation algorithms were studied using the Selox benchmark. Genetic Algorithms, Evolutionary Strategies, Simulated Annealing, Taboo Search and Genetic Algorithms hybridised with Knowledge Discovery procedures were the methods compared. A Design of Experiments search strategy was also exemplified using this benchmark. The main differences regarding the applicability of DoE and Global optimisation techniques are highlighted. Evolutionary strategies, Genetic algorithms, using the sharing procedure, and the Hybrid Genetic algorithms proved to be the most successful in the benchmark optimisation.