摘要
通过动态代谢通量分析方法建立发酵过程模型,提出了一种基于微观代谢信息的发酵过程多目标优化策略,该策略基于所建微观模型,根据动态特性将发酵过程分为菌体生长和产物合成两个阶段,进行特征分析并从微观通量层面分别设计优化目标与约束条件,采用多目标粒子群算法求得最优解。该方法用于青霉素发酵过程底物流加速率和p H的操作轨迹优化,仿真实验结果表明,采用基于微观通量的多目标优化策略能够提高产物终端浓度,表明优化策略的有效性。
A multi-objective micro-scale optimization strategy for fermentation processes was proposed to achieve optimal operation on the basis of a dynamic metabolic flux analysis(DMFA) model. According to different dynamic characteristics, the strategy divided a fermentation process into two stages of cell growth and product synthesis,in which objective functions and constraints were designed from micro metabolic flux and pathways. Multi-objective particle swarm optimization(MOPSO) was employed as key algorithm to find operation trajectory. The strategy was applied to simulation of penicillin fermentation for optimizing acceleration rate of feed stock and p H trajectory. The results showed that terminal product concentration was increased by 3.26% and total feed amount was decreased by 0.6 L, indicating effectiveness of the strategy.
引文
[1]DURAN-VILLALOBOS C A,LENNOX B,LAURI D.Multivariate batch to batch optimisation of fermentation processes incorporating validity constraints[J].Journal of Process Control,2016,46:34-42.
[2]GOLDRICK S,?TEFAN A,LOVETT D,et al.The development of an industrial-scale fed-batch fermentation simulation[J].Journal of Biotechnology,2015,193:70-82.
[3]李慧娟.青霉素发酵过程建模与优化[D].沈阳:东北大学,2010.LI H J.Modeling and optimization for the fermentation process of penicillin[D].Shenyang:Northeastern University,2010.
[4]WU J,WANG J,YU T,et al.An approach to continuous approximation of Pareto front using geometric support vector regression for multi-objective optimization of fermentation process[J].Chinese Journal of Chemical Engineering,2014,22(10):1131-1140.
[5]ZHU Y,YUAN J.A new parameter optimization algorithm of penicillin fermentation model[C]//International Conference on Transportation,Mechanical,and Electrical Engineering.IEEE,2011:1725-1728.
[6]CORTASSA S,CACERES V,BELL L N,et al.From metabolomics to fluxomics:a computational procedure to translate metabolite profiles into metabolic fluxes[J].Biophysical Journal,2015,108(1):163-172.
[7]STEPHANOPOULOS G.Metabolic fluxes and metabolic engineering[J].Metabolic Engineering,1999,1(1):1-11.
[8]LEE S Y,PARK J M,KIM T Y.Application of metabolic flux analysis in metabolic engineering[J].Methods in Enzymology,2011,498:67.
[9]何宁,李寅,陈坚.新型生物絮凝剂REA-11的代谢模型建立与代谢网络分析[J].化工学报,2005,56(4):681-688.HE N,LI Y,CHEN J.Metabolic network modeling and metabolic flux analysis of production of novel bioflocculant REA-11[J].Journal of Chemical Industry&Engineering(China),2005,56(4):681-688.
[10]GAO H J,DU G C,CHEN J.Analysis of metabolic fluxes for hyaluronic acid(HA)production by Streptococcus zooepidemicus[J].World Journal of Microbiology&Biotechnology,2007,22(11):11-11.
[11]FERNANDES S,BASTIN G,WOUWER A V.Metabolic flux analysis of hybridoma cells:underdetermined network and influence of batch and perfusion operating modes[J].IFAC-Papers On Line,2015,48(1):464-469.
[12]VALLINO J J,STEPHANOPOULOS G.Metabolic flux distributions in Corynebacterium glutamicum during growth and lysine overproduction[J].Biotechnology&Bioengineering,2000,67(6):872-885.
[13]VERCAMMEN D,LOGIST F,IMPE J V.Online moving horizon estimation of fluxes in metabolic reaction networks[J].Journal of Process Control,2016,37(6):1-20.
[14]TAKIGUCHI N,SHIMIZU H,SHIOVA S.An on-line physiological state recognition system for the lysine fermentation process based on a metabolic reaction model[J].Biotechnology&Bioengineering,1997,55(1):170.
[15]LEE J S,NISHIKAWA T,MOTTER A E.Why optimal states recruit fewer reactions in metabolic networks[J].Discrete&Continuous Dynamical Systems-Series A(DCDS-A),2017,32(8):2937-2950.
[16]VARMA A,BOESCH B W,PALSSON B O.Biochemical production capabilities of Escherichia coli[J].Biotechnology&Bioengineering,1993,42(1):59.
[17]HOQUE M A,SIDDIQUEE K A Z,SHIMIZU K.Metabolic control analysis of gene-knockout Escherichia coli,based on the inverse flux analysis with experimental verification[J].Biochemical Engineering Journal,2004,19(1):53-59.
[18]MARTINS C P R,SAUTER T,PFAU T.Constraint based modeling going multicellular[J].Frontiers in Molecular Biosciences,2016,3(4):3.
[19]KAUFFMAN K J,PRAKASH P,EDWARDS J S.Advances in flux balance analysis[J].Current Opinion in Biotechnology,2003,14(5):491-496.
[20]TAI M,STEPHANOPOULOS G N.Metabolic engineering:enabling technology for biofuels production[J].Metabolic Engineering,2008,10(6):293-294.
[21]LEIGHTY R W,ANTONIEWICZ M R.Dynamic metabolic flux analysis(DMFA):a framework for determining fluxes at metabolic non-steady state[J].Metabolic Engineering,2011,13(6):745-755.
[22]MARTINEZ V S,BUCHSTEINER M,GRAY P,et al.Dynamic metabolic flux analysis using B-splines to study the effects of temperature shift on CHO cell metabolism[J].Metabolic Engineering Communications,2015,2:46-57.
[23]PRAU?E M T E,SCH?UBLE S,GUTHKE R,et al.Computing the various pathways of penicillin synthesis and their molar yields[J].Biotechnology&Bioengineering,2016,113(1):173-181.
[24]JORGENSEN H,NIELSEN J,VILLADSEN J,et al.Metabolic flux distributions in penicillium chrysogenum during fed-batch cultivations[J].Biotechnology&Bioengineering,1995,46(2):117-131.
[25]KNIGHT J T,ZAHRADKA F T,SINGER D J,et al.Multiobjective particle swarm optimization of a planing craft with uncertainty[J].Journal of Ship Production&Design,2017,30(4):194-200.
[26]KENNEDY J,EBERHART R.Particle Swarm Optimization[M].New York:Springer,2011.
[27]ANOTONIEWICZ M R.Dynamic metabolic flux analysis—tools for probing transient states of metabolic networks[J].Current Opinion Biotechnology,2013,24(6):973-978.
[28]VARMA A,PALSSON B O.Metabolic flux balancing:basic concepts,scientific and practical use[J].Nature Biotechnology,1994,12(10):994-998.
[29]VERCAMMEN D,LOGIST F,IMPE J V.Dynamic estimation of specific fluxes in metabolic networks using non-linear dynamic optimization[J].Computer Aided Chemical Engineering,2014,8(1):132.
[30]GAO Y,ZHAO Z,LIU F.DMFA-based operation model for fermentation processes[J].Computers&Chemical Engineering,2018,109:138-150.
[31]JIN C,ZHANG Y,BALAKRISHNAN S,et al.Local maxima in the likelihood of Gaussian mixture models:structural results and algorithmic consequences[C]//Advances in Neural Information Processing Systems.2016:4116-4124.
[32]YU J,QIN S J.Multiway Gaussian mixture model based multiphase batch process monitoring[J].Industrial&Engineering Chemistry Research,2009,48(18):8585-8594.
[33]ZENG J,XIE L,GAO C,et al.Soft sensor development using non-Gaussian just-in-time modeling[C]//Decision and Control and European Control Conference.IEEE,2011:5868–5873.
[34]KIM S,KANO M,NAKAGAWA H,et al.Estimation of active pharmaceutical ingredients content using locally weighted partial least squares and statistical wavelength selection[J].International Journal of Pharmaceutics,2011,421(2):269-274.
[35]NILASHI M,IBRAHIM O B,ITHNIN N,et al.A multi-criteria collaborative filtering recommender system for the tourism domain using expectation maximization(EM)and PCA-ANFIS[J].Electronic Commerce Research&Applications,2015,14(6):542-562.
[36]耿俊.青霉素发酵过程的模型化研究[D].上海:上海交通大学,2009.GENG J.Research for the modeling of penicillium chrysogenum fed-batch fermentation process[D].Shanghai:Shanghai Jiao Tong University,2009.
[37]JOHNSSON O,ANDERSSON J,LIDEN G,et al.Modelling of the oxygen level response to feed rate perturbations in an industrial scale fermentation process[J].Process Biochemistry,2015,50(4):507-516.
[38]SHI Y,EBERHART R C.Parameter selection in particle swarm optimization[M]//Evolutionary ProgrammingⅦ.Berlin:Springer Berlin Heidelberg,1998:591-600.
[39]HUANG V L,SUGANTHAN P N,LIANG J J.Comprehensive learning particle swarm optimizer for solving multiobjective optimization problems[J].International Journal of Intelligent Systems,2006,21(2):209-226.
[40]裴胜玉,周永权.基于Pareto最优解集的多目标粒子群优化算法[J].计算机工程与科学,2010,32(11):85-88.PEI S Y,ZHOU Y Q.A multi-objective particle swarm algorithm based on the Pareto optimization solution set[J].Computer Engineering&Science,2010,32(11):85-88.
[41]LIU Y.Modelling of the penicillin fermentation process via LS-SVM based on Pensim simulator[J].Chemical Reaction Engineering&Technology,2006,22(3):252-258.
[42]HUANG J,JI G,ZHU Y,et al.Identification of multi-model LPV models with two scheduling variables[J].Journal of Process Control,2012,22(7):1198-1208.
[43]SEKAR B S,SEOL E,RAJ S M,et al.Co-production of hydrogen and ethanol by pfk A-deficient Escherichia coli,with activated pentose-phosphate pathway:reduction of pyruvate accumulation[J].Biotechnology for Biofuels,2016,9(1):95.