Validation of uniform designs use in combination with neural networks in bioprocess

• Grant number: 14/03883-3
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: May 01, 2014
• End date: July 29, 2014
• Field of knowledge: Engineering - Chemical Engineering
• Principal Investigator: Eutimio Gustavo Fernández Núñez
• Grantee: Laís Hara Buenno
• Host Institution: Faculdade de Ciências e Letras (FCL-ASSIS). Universidade Estadual Paulista (UNESP). Campus de Assis. Assis , SP, Brazil

Abstract

Quality by Design (QbD) and Process Analytical Technology (PAT) initiatives proposed by Food and Drug Administration (USFDA), since 2002, have stimulated pharmaceutical industry worldwide to develop process with high level of scientific understanding of the relationship between operation parameters and product and process critical quality attributes, from early stages of technological development, as well as rigorous control over individual operations in established industrial processes. Thus, the paradigm of quality by inspection is changed for a new one that reduces costs by faults in the manufacture of pharmaceuticals. Among the incorporated tools to put these initiatives into practice, statistical designs of experiments and artificial neural networks are included. This project aims to validate the combined use of these two techniques to develop a bioprocess (production of rabies vaccine). Once the critical parameters that influence viral multiplication were defined, a set of combinations of different values within their respective ranges with technological significance, will be generated and evaluated experimentally according to a uniform experimental design (relatively smaller number of combinations when compared with traditional factorial designs) and an supervised artificial neural network, "multilayer perceptron" model, with backpropagation learning algorithm, will be used as correlation technique. This combination of methods can reduce the time and resources in the stages of research and development of biopharmaceutical organizations, issue of high importance in the current context of this business sector.

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

BUENNO, LAIS HARA; ROCHA, JOSE CELSO; LEME, JACI; CARICATI, CELSO PEREIRA; TONSO, ALDO; FERNANDEZ NUNEZ, EUTIMIO GUSTAVO. Use of Uniform Designs in Combination with Neural Networks for Viral Infection Process Development. BIOTECHNOLOGY PROGRESS, v. 31, n. 2, p. 532-540, MAR-APR 2015. (14/03883-3)