Deep neural network method for the prediction of xylitol production

• Main Authors: Yousoff, S. N. M., Baharin, A., Abdullah, A.
• Format: Article
• Language: English
• Published: Institute of Advanced Engineering and Science 2017
• Subjects: QA Mathematics
• Online Access: http://eprints.utm.my/id/eprint/74892/1/SitiNoorainMohmad_DeepNeuralNetworkMethod.pdf; http://eprints.utm.my/id/eprint/74892/; https://www.scopus.com/inward/record.uri?eid=2-s2.0-85016927966&doi=10.11591%2fijeecs.v5.i3.pp691-696&partnerID=40&md5=2727a069bd76a32f8b6fd4b2b2f7bbfe
• Institution: Universiti Teknologi Malaysia

Bio-based chemical products such as xylitol have achieved remarkable attentions both in pharmaceutical and food industries due to their several advantages such as sugar substitute that can help diabetic patients and help in preventing tooth decay problem. To produce xylitol, recently, microbial host such as E. Coli often used as it is predicted that E. Coli can produce high level of xylitol. Therefore, metabolic engineering need to be done towards E. Coli and powerful tools are needed to manipulate, simulate and analyse the E. Coli metabolic pathway. Artificial intelligence methods such as deep neural network offer an efficient and powerful approach to be used to analyse the xylitol production value and at the same time to predict which genes and pathway that give biggest effect in the process to produce xylitol in E. Coli. Results show that, with an absence of genes pgi, tkt and tala, xylitol production can be boosted up to the higher level.