| CPC C12M 47/10 (2013.01) [C07K 1/16 (2013.01); C12M 29/04 (2013.01); C12M 29/10 (2013.01); C12M 37/00 (2013.01); C12M 47/12 (2013.01); C12P 21/00 (2013.01)] | 20 Claims |
|
1. A system to produce one or more biotherapeutics, the system comprising:
a production facility that includes:
a first modular cleanroom comprising a first plurality of pieces of equipment to produce an effluent that includes a biotherapeutic, the first plurality of pieces of equipment including a bioreactor;
a second modular cleanroom comprising a second plurality of pieces of equipment to purify the effluent produced by the first plurality of pieces of equipment, the second plurality of pieces of equipment including at least a filter system;
a staging area including a plurality of storage containers, a first portion of the plurality of storage containers storing cell culture media and cell growth material for the bioreactor and a second portion of the plurality of storage containers storing buffer solution for at least one piece of equipment included in the first plurality of pieces of equipment and for at least one piece of equipment included in the second plurality of pieces of equipment;
one or more processors; and
one or more computer-readable storage media storing computer-readable instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
obtaining a first set of data indicating a first plurality of conditions of the bioreactor, the first set of data being obtained via at least one sensor of a plurality of sensors of the bioreactor or one or more external assays that measure conditions at different points in time during operation of the bioreactor;
analyzing the first set of data utilizing one or more inferential modeling techniques to determine one or more process variables that have at least a first threshold impact on a productivity metric related to a bioreactor effluent;
analyzing the first set of data utilizing the one or more inferential modeling techniques to determine one or more control variables that have at least a second threshold impact on the one or more process variables;
generating a model that includes variables corresponding to the one or more process variables and the one or more control variables, wherein the model predicts the productivity metric;
obtaining process data from at least one of one or more first pieces of equipment of the first plurality of pieces of equipment or one or more second pieces of equipment of the second plurality of pieces of equipment;
applying the process data to the model to determine values of the productivity metric:
responsive to determining that at least a portion of the values of the productivity metric are outside of a threshold range of values, determining, by applying one or more machine learning techniques to the process data, at least one process variable to modify to change the productivity metric;
determining at least one control variable to modify to cause additional values of the at least one process variable to change;
sending control signals to one or more pieces of equipment, the control signals indicating control settings of the one or more pieces of equipment to modify the at least one control variable in a manner that modifies the productivity metric; and
causing operation of the one or more pieces of equipment to change based on the control signals.
|