Method for design and analysis on multi-column continuous chromatography

Patent metadata

Field	Value
Publication number	US-12540928-B2
Application number	US-201917623403-A
Country	US
Kind code	B2
Filing date	Sep 5, 2019
Priority date	Jun 28, 2019
Publication date	Feb 3, 2026
Grant date	Feb 3, 2026

Abstract

The present invention discloses a method for realizing multi-column continuous chromatography design and analysis based on a chromatography model, and a method for realizing multi-column continuous chromatography design and analysis based on an artificial neural network. The method based on the chromatography model includes the following steps: step 101 , experimental breakthrough curve fitting: performing fitting using a chromatography model to obtain model parameters; step 102 : breakthrough curve prediction: substituting the model parameters into the chromatography model to obtain a breakthrough curve under different operation conditions; step 103 , process analysis of continuous chromatography: substituting the predicted breakthrough curve and the continuous chromatography operation parameters into a continuous chromatography model to obtain performance indexes such as process productivity and resin capacity utilization; and step 104 , operation space optimization of continuous chromatography: obtaining the operation space of the continuous chromatography design parameters based on a specific separation target. The method based on the artificial neutral network completes the respective steps above by replacing the chromatography model with artificial neural network.

First claim

The invention claimed is:

1 . A method for realizing multi-column continuous chromatography design and analysis based on a chromatography model for protein separation, comprising the following steps:
step 101 , experimental breakthrough curve fitting: conducting a single-column breakthrough experiment using Protein A resin and a protein feed solution at a flow rate of 1 mL/min and a protein concentration of 2 mg/mL, wherein feeding is stopped at 95% breakthrough concentration to obtain an experimental breakthrough curve; substituting the experimental breakthrough curve and chromatography operation parameters into a chromatography mechanistic model that incorporates an axial diffusion coefficient, liquid-film mass transfer coefficient, solid-phase and liquid-phase mass transfer coefficients, saturated adsorption capacity, and equilibrium desorption constant; fitting the breakthrough curve by calculating a simulated breakthrough curve using an orthogonal collocation method and iteratively optimizing model parameters by minimizing a root mean square deviation (RMSD) objective function, thereby obtaining fitted mechanistic model parameters; step 102 , breakthrough curve prediction: defining a chromatography operating range, wherein a flow rate is set between 0.33 mL/min and 3 mL/min and a protein concentration is set between 0.5 mg/mL and 5 mg/mL; substituting the fitted mechanistic model parameters obtained in step 101 and the defined chromatography operation parameters into the chromatography mechanistic model to calculate one-column breakthrough curves and twin-column in-series breakthrough curves at different flow rates and protein concentrations, wherein the protein concentration profile of the one-column breakthrough curve serves as a feed concentration input for the second column in the twin-column configuration; step 103 , process analysis of continuous chromatography: substituting the breakthrough curves predicted in step 102 and basic continuous chromatography operation parameters comprising an interconnected feeding flow rate, a switching points, a washing flow rate, a washing column volume, a safety factor, a total time of elution, cleaning and regeneration, into a continuous chromatography model; determining design parameters comprising a disconnected feeding flow rate, an interconnected feeding time and a waiting time; analyzing influence of operation parameter variations on performance indexes comprising process productivity and resin capacity utilization for the multi-column continuous chromatography process; and step 104 , operation space optimization of the continuous chromatography: determining a suitable process based on a specific separation target and operational requirements of productivity of at least 40 g/L/h and resin capacity utilization of at least 80%; generating process productivity and resin capacity utilization matrices by varying switching point parameters ranging from 0.1 to 0.9 and residence times ranging from 0.33 min to 3 min; performing linear interpolation on the matrices to obtain process productivity and resin capacity utilization contour diagrams; identifying an optimal operation space by locating intersections where both process productivity and resin capacity utilization satisfy the operational requirements, thereby defining the optimized continuous chromatography design parameters.

2 . The method for realizing multi-column continuous chromatography design and analysis based on a chromatography model according to claim 1 , wherein the experimental breakthrough curve fitting in the step 101 further comprises the following steps:
substituting the experimental operation parameters of the breakthrough curve and initial values of the model parameters into the chromatography mechanistic model to calculate a breakthrough curve, and comparing the calculated result with the breakthrough curve obtained by experiments; and changing the chromatography model parameters to minimize a root mean square error (RMSD) thereof to obtain chromatography model parameters, thus realizing the breakthrough curve fitting.

3 . The method for realizing multi-column continuous chromatography design and analysis based on a chromatography model according to claim 1 , wherein the breakthrough curve prediction in the step 102 further comprises the following steps:
setting a flow rate and a protein concentration range of the chromatography operation, and generating a chromatography operation parameter matrix within the range and merging with the mechanistic model parameters obtained in the step 101 , and substituting into the chromatography model for calculation, and performing predicting to obtain the one-column and twin-column in-series breakthrough curves at different flow rates and different protein concentrations.

4 . The method for realizing multi-column continuous chromatography design and analysis based on a chromatography model according to claim 1 , wherein the process analysis of the continuous chromatography comprises the following steps:
substituting the predicted breakthrough curve and the basic operation parameters of the continuous chromatography into the continuous chromatography model to obtain process design parameters and a process scheduling of the continuous chromatography; substituting the obtained process design parameters and the process scheduling of the continuous chromatography into an evaluation model of the continuous chromatography, and performing calculating to obtain process productivity and resin capacity utilization of the multi-column continuous chromatography.

5 . The method for realizing multi-column continuous chromatography design and analysis based on a chromatography model according to claim 1 , wherein the operation space of the continuous chromatography comprises the following steps:
generating a parameter matrix based on a design parameter range of the continuous chromatography, calculating process productivities for all the parameter points within the parameter matrix to obtain a process productivity matrix; and performing linear interpolation on the process productivity matrix and drawing a process productivity contour diagram under different operation conditions for the process analysis and optimization of the continuous chromatography; generating a parameter matrix based on a design parameter range of the continuous chromatography, calculating resin capacity utilization to all the parameter points within the parameter matrix to obtain a resin capacity utilization matrix; and performing linear interpolation on the resin capacity utilization matrix and drawing a resin capacity utilization contour diagram under different operation conditions for the process analysis and optimization of the continuous chromatography; calculating a continuous chromatography design parameter range satisfying the separation target respectively in the process productivity contour diagram and in the resin capacity utilization contour diagram based on the specific separation target, and performing superposition on design parameter regions of the two diagrams to obtain design parameters of continuous chromatography satisfying the requirements of process productivity and resin capacity utilization simultaneously, and calculating operation parameters and process scheduling of the continuous chromatography process.

6 . The method for realizing multi-column continuous chromatography design and analysis based on a chromatography model according to claim 1 , wherein the chromatography model is a General Rate Model taking parallel diffusion into consideration.

7 . The method for realizing multi-column continuous chromatography design and analysis based on a chromatography model according to claim 1 , wherein:
the c

Assignees

• Univ Zhejiang

Inventors

• Lin Dong Qiang
• Shi Ce
• Yao Shan Jing

Classifications

• G06F17/00

  Digital computing or data processing equipment or methods, specially adapted for specific functions (information retrieval, database structures or file system structures therefor G06F16/00) · CPC title

• B01D15/1871

  placed in series · CPC title

• G01N30/8693 Primary

  Models, e.g. prediction of retention times, method development and validation · CPC title

• G01N30/8658 Primary

  Optimising operation parameters · CPC title

• G06N3/0499

  Feedforward networks · CPC title