# US 12,182,482 B2 · Method for predicting outcome of an modelling of a process in a bioreactor

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,182,482 B2
- Document ID: US12182482B2
- Status: Granted
- Kind: B2
- Application number: 16/620,940
- Filed: 2018-06-18
- Published: 2020-06-25
- Granted: 2024-12-31
- Earliest claimed: 2017-06-16
- Assignees: Cytiva Sweden AB
- Applicants: Cytiva Sweden AB
- Inventors: Krishna Kumar Swaminathan; Sridhar Dasaratha; Sittal George; Vinay Bhaskar Jammu; Alok Singh Chauhan; Nagaraju Konduru; Neelima Boddapati; Ankita Bhatia; Mithilesh Mohanty; Harsha Aeron; Andreas Castan
- Web page: https://patentgrep.com/patent/US12182482B2

## CPC

- C12M 41/48: {Automatic or computerized control (automatic analysis G01N35/00)}
- G06F 30/20: Design optimisation, verification or simulation (optimisation, verification or simulation of circuit designs G06F30/30)
- G05B 13/04: involving the use of models or simulators
- G05B 17/02: electric
- C12M: APPARATUS FOR ENZYMOLOGY OR MICROBIOLOGY; {APPARATUS FOR CULTURING MICROORGANISMS FOR PRODUCING BIOMASS, FOR GROWING CELLS OR FOR OBTAINING FERMENTATION OR METABOLIC PRODUCTS, i.e. BIOREACTORS OR FERMENTERS}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G)

## Abstract

The present invention relates to a method for predicting outcome of a process used for manufacturing a sample in a bioreactor, the process belonging to a category. The method comprises selecting (51) a process model based on the category; accessing (53) historic data related to past process runs for manufacturing the sample; accessing (53) current data obtained (54) from a current process run of the process. The obtained current data, which is based on the selected process model, comprises: process strategy data, bioreactor instrument data, data from online sensors and/or data from offline sensors. The method further comprises predicting (62) an outcome of at least one selected parameter of the current process run for manufacturing the sample based on the accessed historic data and current data. The present invention also relates to a method for modelling a process and a control system (10) for controlling a process.

## Claims

### Claim 1

A method for predicting outcome of a process used for manufacturing a sample in a bioreactor, the process belonging to a multi-level knowledge tree, wherein the method comprises:
selecting a process model based on the multi-level knowledge tree, wherein each level of the knowledge tree corresponds to a category associated with a type of process;
accessing historic data related to past process runs for manufacturing the sample;
accessing current data obtained from a current process run of the process, wherein the obtained current data, which is based on the selected process model, comprises: process strategy data, bioreactor instrument data, data from online sensors and/or data from offline sensors;
subselecting historic data of previous process runs similar to the current process run from the accessed historic data;
predicting an outcome of at least one selected parameter selected from cell viability, cell count, product titre, and product quality of the current process run for manufacturing the sample based on the subselected historic data and current data;
automatically determining an updated model for predicting outcomes of future process runs, wherein the updated model is determined based at least in part on:
the process model based on the multi-level knowledge tree; and
additional data from the current process run; and
applying the updated model to future process runs of other samples.

### Claim 2 (depends on 1)

The method according to claim 1, wherein the method further comprises consolidating historic data, current data and data related to the process model in a database.

### Claim 3 (depends on 1)

The method according to claim 1, wherein obtaining current data further comprises:
selecting parameters based on the base model used, and
reading data of selected parameters.

### Claim 4 (depends on 1)

The method according to claim 1, wherein the method further comprises handling of missing data in the current data obtained from the current process run.

### Claim 5 (depends on 4)

The method according to claim 4, wherein handling of missing data further comprises:
replacing missing data values with imputed values, or
removing data with missing data values.

### Claim 6 (depends on 1)

The method according to claim 1, wherein the method further comprises adapting the process model in real-time based on the current data.

### Claim 7 (depends on 1)

The method according to claim 1, wherein the updated model is determined in response to completion of the current process run.

### Claim 8 (depends on 6)

The method according to claim 6, wherein the process used in the current process run is determined to belong to a new category and the step of adapting the process run further comprises:
assigning the process to the new category; and
storing the process model as a new process model.

### Claim 9 (depends on 1)

The method according to claim 1, wherein the step of predicting the outcome of at least one selected parameter further comprises:
predicting a forecast of the at least one parameter; and/or
predicting a forecast of anomalies; and/or
determining and recommending actions to obtain improved conditions for the current process run.

### Claim 10 (depends on 1)

A non-transitory computer-readable storage medium carrying a computer program predicting outcome of a current process run for manufacturing a sample in a bioreactor according to claim 1.

### Claim 11

A method for modelling of a process used for manufacturing a sample in a bioreactor, the process belonging to a multi-level knowledge tree, wherein the method comprises:
selecting a process model based on the multi-level knowledge tree, wherein each level of the knowledge tree corresponds to a category associated with a type of process;
accessing historic data related to past process runs for manufacturing the sample;
accessing current data obtained from a current process run of the process, wherein the current data, which is based on the selected process model, comprises: process strategy data, bioreactor instrument data, data from online sensors and/or data from offline sensors;
monitoring at least one parameter selected from cell viability, cell count, product titre, and product quality of the current process run for manufacturing the sample;
subselecting historic data of previous process runs similar to the current process run from the accessed historic data;
automatically determining an updated model for predicting outcomes of future process runs, wherein the predicted outcome of future process runs comprises a prediction of the at least one parameter, and wherein the updated model is determined based at least in part on:
the process model based on the multi-level knowledge tree; and
additional data from the current process run; and
applying the updated model to future process runs of other samples.

### Claim 12 (depends on 11)

The method according to claim 11, wherein the method further comprises consolidating historic data, current data and data related to the process model in a database.

### Claim 13 (depends on 11)

The method according to claim 11, wherein obtaining current data further comprises:
selecting parameters based on the base model used, and
reading data of selected parameters.

### Claim 14 (depends on 11)

The method according to claim 11, wherein the method further comprises handling of missing data in the current data obtained from the current process run.

### Claim 15 (depends on 14)

The method according to claim 14, wherein handling of missing data further comprises:
replacing missing data values with imputed values, or
removing data with missing data values.

### Claim 16 (depends on 11)

The method according to claim 11, wherein the step of adapting the process model further comprises updating the process model for the category.

### Claim 17 (depends on 11)

The method according to claim 11, wherein the process used in the current process run is determined to belong to a new category and the step of adapting the process model further comprises creating a new process model by:
assigning the process to the new category; and
storing the process model as a new process model.

### Claim 18 (depends on 11)

A non-transitory computer-readable storage medium carrying a computer program for modelling of a process used for manufacturing a sample in a bioreactor according to claim 11.

### Claim 19

A control system for controlling a process used for manufacturing a sample in a bioreactor, the process belonging to a multi-level knowledge tree, wherein the control system is configured to model the process and is further configured to:
select a process model based on the multi-level knowledge tree, wherein each level of the knowledge tree corresponds to a category associated with a type of process;
access historic data related to past process runs for manufacturing the sample;
access current data obtained from a current process run of the process, wherein the obtained data, which is based on the selected process model, comprises: process strategy data, bioreactor instrument data, data from online sensors and/or data from offline sensors;
subselect historic data of previous process runs similar to the current process run from the accessed historic data;
predict an outcome of at least one selected parameter selected from cell viability, cell count, product titre, and product quality of the current process run for manufacturing the sample;
control the process used for manufacturing the sample in a bioreactor based on the predicted outcome of the at least one selected parameter of the current process run;
automatically determine an updated model for predicting outcomes of future process runs, wherein the updated model is determined based at least in part on:
the process model based on the multi-level knowledge tree; and
additional data from the current process run; and
applying the updated model to future process runs of other samples.

### Claim 20 (depends on 19)

The control system according to claim 19, wherein the sample is introduced into a separating system configured to purify the sample, and the control system is further configured to access data from the separating system and to improve the prediction based on the data from the separating system.

### Claim 21 (depends on 19)

The control system according to claim 19, wherein the control system is configured to have access to a database with consolidated data comprising historic data, current data and data related to the process model.

### Claim 22

A method for predicting forecasts for a feature in a process used for manufacturing a sample in a bioreactor during a process run, the process belonging to a multi-level knowledge tree, wherein each level of the knowledge-tree corresponds to a category associated with a type of process, values related to the feature being continuously measured during a current process run, wherein the method comprises:
creating a model for the current process run based on a selection of historic data;
after a time period during the current process run, automatically selecting best fitting historic data related to the values related to the feature being continuously measured during the current process run by subselecting historic data of previous process runs similar to the current process run, and updating the model to produce an updated model, wherein the updated model is determined based on the best fitting historic data and the model;
revising forecasts during the current process run of at least one process parameter selected from cell viability, cell count, product titre, and product quality from the updated model based on calculated errors between measured values and the updated model; and
applying the updated model for future process runs of other samples.

### Claim 23 (depends on 22)

The method according to claim 22, wherein the method further comprises:
performing residual error correction on the revised forecasts using metabolite information.

### Claim 24 (depends on 22)

The method according to claim 22, further comprising:
setting a set of conditions for the model, and
obtaining an amount of historic data from previous process runs to form the selection of historic data used to create the model.

### Claim 25 (depends on 24)

The method according to claim 24, wherein the method further comprises:
controlling if the amount of historic data obtained from previous runs is in a predetermined interval, and
updating conditions for the model if the amount of historic data is outside the predetermined interval and repeat the step of obtaining an amount of historic data, or proceed to the step of creating a model if the amount of historic data is within the predetermined interval.

### Claim 26 (depends on 25)

The method according to claim 25, wherein the predetermined interval is selected to be historic data from at least ten previous process runs.

### Claim 27 (depends on 25)

The method according to claim 25, wherein the predetermined interval is selected to be historic data from not more than one hundred previous process runs.

### Claim 28 (depends on 22)

The method according to claim 22, wherein the method further comprises:
storing data from the current process run as historic data for future runs.

### Claim 29 (depends on 1)

The method according to claim 1, wherein the at least one process parameter is cell viability or cell count.

### Claim 30 (depends on 11)

The method according to claim 11, wherein the at least one process parameter is cell viability or cell count.

### Claim 31 (depends on 19)

The control system according to claim 19, wherein the at least one process parameter is cell viability or cell count.

### Claim 32 (depends on 22)

The method according to claim 22, wherein the at least one process parameter is cell viability or cell count.