US 11,655,444 B2
Methods, devices, and computer program products for standardizing a fermentation process
Steven B. Haase, Pittsboro, NC (US); Adam R. Leman, Chapel Hill, NC (US); David S. Morris, Durham, NC (US); and Ashlee M. Valente, Cary, NC (US)
Assigned to PRECISION FERMENTATION, INC., Durham, NC (US)
Filed by PRECISION FERMENTATION, INC., Durham, NC (US)
Filed on Sep. 26, 2018, as Appl. No. 16/142,736.
Claims priority of provisional application 62/564,816, filed on Sep. 28, 2017.
Prior Publication US 2019/0093065 A1, Mar. 28, 2019
Int. Cl. C12M 1/36 (2006.01); C12N 1/16 (2006.01); C12N 1/20 (2006.01); C12Q 1/06 (2006.01); C12N 15/10 (2006.01); C12N 1/12 (2006.01); G16B 20/00 (2019.01); G16B 25/00 (2019.01); G16B 50/00 (2019.01); C12C 11/00 (2006.01); G16B 25/10 (2019.01); C12Q 1/6809 (2018.01); C12N 1/18 (2006.01); C12Q 1/68 (2018.01); C12Q 1/6806 (2018.01); G16B 5/30 (2019.01); C12R 1/225 (2006.01); C12R 1/865 (2006.01)
CPC C12M 41/48 (2013.01) [C12C 11/00 (2013.01); C12N 1/12 (2013.01); C12N 1/16 (2013.01); C12N 1/18 (2013.01); C12N 1/185 (2021.05); C12N 1/20 (2013.01); C12N 1/205 (2021.05); C12N 15/1003 (2013.01); C12Q 1/06 (2013.01); C12Q 1/68 (2013.01); C12Q 1/6806 (2013.01); C12Q 1/6809 (2013.01); G16B 5/30 (2019.02); G16B 20/00 (2019.02); G16B 25/00 (2019.02); G16B 25/10 (2019.02); G16B 50/00 (2019.02); C12R 2001/225 (2021.05); C12R 2001/865 (2021.05)] 6 Claims
 
5. A method providing a technical solution to the technical problem of standardizing a selected fermentation process by a fermentation organism in a fermentation substrate, the method comprising:
(I) first, constructing a baseline database for the selected fermentation process by the fermentation organism in the fermentation substrate by
(a) initiating a first instance of the selected fermentation process by the fermentation organism in the fermentation substrate and obtaining, at each respective time point of a plurality of predefined time points defined from the beginning of the initiated first instance of the fermentation process, a respective fluidic sample,
(b) measuring, using each respective fluidic sample for the first instance, one or more physical parameters for the respective fluidic sample at the corresponding respective time point,
(c) determining one or more physical parameter values for the first instance based on the measuring for the first instance, the one or more physical parameter values including values at a point in time and values representing a rate of change,
(d) measuring, using each respective fluidic sample for the first instance, a transcriptome of the fermentation organism at the corresponding respective time point, such measuring comprising
(i) isolating RNA from the fermentation substrate of the respective fluidic sample,
(ii) purifying the RNA isolated from the fermentation substrate of the respective fluidic sample, and
(iii) measuring the RNA;
(e) determining gene expression data for the selected fermentation process based on the obtained measurements by
(i) filtering determined physical parameter and gene expression data to generate a first dataset which only includes dynamic physical parameter values and dynamic gene expression values,
(ii) computationally normalizing dynamic physical parameter values and dynamic gene expression values of the first dataset to generate a normalized dataset,
(iii) determining one or more possible regulators by identifying dynamic gene expression values of the normalized dataset that correspond to transcription factors, and
(iv) comparing normalized dynamic physical parameter values and normalized dynamic gene expression values of the normalized dataset as targets to each determined possible regulator by
(A) generating a regulation function for each possible regulator-target relationship, each regulation function defining a relationship between one of the determined possible regulators and a downstream gene target corresponding to one of the normalized dynamic gene expression values or a chemical change target corresponding to one of the normalized dynamic physical parameter values,
(B) calculating, for each regulator-target relationship, a score representing a fit of the corresponding possible regulator to the corresponding target,
(C) ranking each regulation-target relationship based on the calculated scores, and assigning a confidence value to each regulator-target relationship,
(D) determining a confidence threshold based at least in part on data density, and
(E) constructing a regulatory network based on the ranked regulator-target relationships and the confidence threshold,
(f) constructing, based on the ranked regulator-target relationships and the constructed regulatory network, the baseline database for the selected fermentation process that specifies
(i) one or more condition sets each comprising a preferred value or range of values, at one or more respective time points of the plurality of predefined time points, for one or more physical parameters that have been determined based on the ranked regulator-target relationships and the constructed regulatory network to correspond to one or more regulatory genes of the fermentation organism,
(ii) for each condition set, one or more remediation actions determined, based on the ranked regulator-target relationships and the constructed regulatory network, to increase or decrease the expression of one or more regulatory genes of the fermentation organism determined based on the ranked regulator-target relationships and the constructed regulatory network to correspond to the respective physical parameter;
(II) initiating, in a fermentation vessel, a standardized instance of the selected fermentation process by the fermentation organism in the fermentation substrate by
(a) initiating a second instance of the selected fermentation process by the fermentation organism in the fermentation substrate,
(b) automatically, at each respective time point of the plurality of predefined time points defined from the beginning of the initiated first instance of the fermentation process,
(i) obtaining a respective fluidic sample,
(ii) measuring, using the respective fluidic sample for the second instance, one or more physical parameters for the respective fluidic sample at the corresponding respective time point,
(iii) determining one or more physical parameter values for the second instance based on the measuring for the second instance, the one or more physical parameter values including values at a point in time and values representing a rate of change,
(iv) comparing determined physical parameter values for the second instance to preferred values and ranges of values specified in condition sets of the baseline database,
(c) automatically identifying, as a result of comparing at a certain one of the time points determined physical parameter values for the second instance to preferred values and ranges of values specified in condition sets of the baseline database, a first physical parameter value for a first physical parameter which falls outside of a preferred range of values specified for the first physical parameter by a first condition set of the baseline database,
(d) automatically determining, via lookup in the baseline database, a first remediation action determined, based on the ranked regulator-target relationships and the constructed regulatory network, to increase or decrease the expression of one or more regulatory genes of the fermentation organism determined based on the ranked regulator-target relationships and the constructed regulatory network to correspond to the first physical parameter, the first remediation action comprising modifying a specified first fermentation condition,
(e) effecting modification of the specified first fermentation condition to affect the expression of the determined first regulatory gene,
wherein the one or more physical parameters comprises density of the fermentation substrate and the first regulatory gene is ADH1 or any homologue thereof.