Controlling the product selectivity in the conversion of methanol to the feedstock for phenol production.

This work investigated a novel nonpetroleum-based catalytic process of methanol to phenol. The idea was to convert methanol to produce a main product stream having a molar ratio of propylene to benzene/toluene of unity along with relatively higher-value products including -xylene and alkenes. Such a product mix would be ideal for the manufacturing of phenol.

Design and Application of a Variable Selection Method for Multilayer Perceptron Neural Network With LASSO.

In this paper, a novel variable selection method for neural network that can be applied to describe nonlinear industrial processes is developed. The proposed method is an iterative two-step approach. First, a multilayer perceptron is constructed.

On the local optimal solutions of metabolic regulatory networks using information guided genetic algorithm approach and clustering analysis.

Biological information generated by high-throughput technology has made systems approach feasible for many biological problems. By this approach, optimization of metabolic pathway has been successfully applied in the amino acid production. However, in this technique, gene modifications of metabolic control architecture as well as enzyme expression levels are coupled and result in a mixed integer nonlinear programming problem.

Biological network mapping and source signal deduction.

Motivation: Many biological networks, including transcriptional regulation, metabolism, and the absorbance spectra of metabolite mixtures, can be represented in a bipartite fashion. Key to understanding these bipartite networks are the network architecture and governing source signals. Such information is often implicitly imbedded in the data.