Prediction of bioprocess production using deep neural network method

• Main Authors: Baharin, A., Abdullah, A., Yousoff, S. N. M.
• Format: Article
• Language: English
• Published: Universitas Ahmad Dahlan 2017
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://eprints.utm.my/id/eprint/75671/1/AmirahBaharin_PredictionofBioprocessProductionUsingDeep.pdf; http://eprints.utm.my/id/eprint/75671/; https://www.scopus.com/inward/record.uri?eid=2-s2.0-85020228565&doi=10.12928%2fTELKOMNIKA.v15i2.6124&partnerID=40&md5=b31d58bc81ccceba28600aa6fd1399fb

Deep learning enhanced the state-of-the-art methods in genomics allows it to be used in analysing the biological data with high prediction. The training process of neural network with several hidden layers which has been facilitated by deep learning has been subjected into increased interest in achieving remarkable results in various fields. Thus, the extraction of bioprocess production can be implemented by pathway prediction in genomic metabolic network in eschericia coli. As metabolic engineering involves the manipulation of genes which have the potential to increase the yield of metabolite production. A mathematical model of this network is the foundation for the development of computational procedure that directs genetic manipulations that would eventually lead to optimized bioprocess production. Due to the ability of deep learning to be well suited in terms of genomics, modelling for biological network can be implemented. Each layer reveal the insight of biological network which enable pathway analysis to be implemented in order to extract the target bioprocess production. In this study, deep neural network has been to identify any set of gene deletion models that offers optimal results in xylitol production and its growth yield.