Microbial strain improvement by a HTP genomic engineering platform
The present disclosure provides a HTP microbial genomic engineering platform that is computationally driven and integrates molecular biology, automation, and advanced machine learning protocols. This integrative platform utilizes a suite of HTP molecular tool sets to create HTP genetic design libraries, which are derived from, inter alia, scientific insight and iterative pattern recognition. The HTP genomic engineering platform described herein is microbial strain host agnostic and therefore can be implemented across taxa. Furthermore, the disclosed platform can be implemented to modulate or improve any microbial host parameter of interest.
1. A high-throughput method for engineering a host cell to have improved phenotypic performance, comprising:
a. accessing a training data set containing one or more genetic alteration input variables and one or more measured phenotypic performance output variables,
i. wherein the one or more genetic alteration input variables represent one or more genetic alterations that have been introduced into a host cell, and
ii. wherein the one or more measured phenotypic performance output variables represent one or more phenotypic performance measurements that are associated with the one or more introduced genetic alterations;
b. developing a predictive machine learning model that is populated with the training data set;
c. generating, in silico, a pool of design candidate host cells incorporating the one or more genetic alterations;
d. utilizing the predictive machine learning model to predict the expected phenotypic performance of each member of the pool of design candidate host cells,
i. wherein at least one design candidate host cell comprises a consolidated combination of genetic alterations from among the genetic alterations of step (a), in a genomic sequence, said combination being uncharacterized for improving the phenotypic performance at the time that step (d) is carried out;
ii. wherein the expected phenotypic performance predicted by the machine learning model is based upon the introduced genetic alterations and their associated phenotypic performance measurements of step (a);
e. selecting a subset of the design candidate host cells based upon their predicted phenotypic performance;
f. manufacturing host cells from the subset of the design candidate host cells to thereby create engineered host cells;
g. measuring, in an in vitro assay, the phenotypic performance of the engineered host cells; and
h. adding to the training data set of (a):
i. one or more genetic alteration input variables representing one or more genetic alterations that were introduced into the engineered host cells, and
ii. one or more measured phenotypic performance output variables representing the phenotypic performance measurements of the engineered host cells.
2. The method of claim 1 , wherein (a)-(h) are repeated until an engineered host cell exhibits a desired level of improved phenotypic performance.
3. The method of claim 1 , wherein the predictive machine learning model incorporates epistatic effects.
4. The method of claim 1 , wherein the predictive machine learning model incorporates at least one of the following: linear regression, kernel ridge regression, logistic regression, neural networks, support vector machines (SVMs), decision trees, hidden Markov models, Bayesian networks, a Gram-Schmidt process, reinforcement-based learning, cluster-based learning, hierarchical clustering, genetic algorithms, and combinations thereof.
5. The method of claim 1 , wherein the predictive machine learning model is supervised, semi-supervised, or unsupervised.
6. The method of claim 1 , wherein the one or more genetic alterations comprise at least one genetic alteration selected from the group consisting of: a single nucleotide polymorphism, nucleotide sequence insertion, nucleotide sequence deletion, and nucleotide sequence replacements.
7. The method of claim 1 , wherein the one or more genetic alterations comprise one or more heterologous promoters from a promoter ladder operably linked to an endogenous target gene.
8. The method of claim 1 , wherein the improved phenotypic performance is increased or more efficient production of a product of interest, said product of interest selected from the group consisting of: a small molecule, enzyme, protein, peptide, amino acid, organic acid, synthetic compound, fuel, alcohol, primary extracellular metabolite, secondary extracellular metabolite, intracellular component molecule, and combinations thereof.
9. A high-throughput method for engineering a microbial strain to have improved phenotypic performance, comprising:
a. accessing a training data set containing one or more genetic alteration input variables and one or more measured phenotypic performance output variables,
i. wherein the one or more genetic alteration input variables represent one or more genetic alterations that have been introduced into a microbial strain, and
ii. wherein the one or more measured phenotypic performance output variables represent one or more phenotypic performance measurements that are associated with the one or more introduced genetic alterations;
b. developing a predictive machine learning model that is populated with the training data set;
c. generating, in silico, a pool of design candidate microbial strains incorporating the one or more genetic alterations;
d. utilizing the predictive machine learning model to predict the expected phenotypic performance of each member of the pool of design candidate microbial strains,
i. wherein at least one design candidate microbial strain comprises a consolidated combination of genetic alterations from among the genetic alterations of step (a), in a genomic sequence, said combination being uncharacterized for improving the phenotypic performance at the time that step (d) is carried out;
ii. wherein the expected phenotypic performance predicted by the machine learning model is based upon the introduced genetic alterations and their associated phenotypic performance measurements of step (a);
e. selecting a subset of the design candidate microbial strains based upon their predicted phenotypic performance;
f. manufacturing microbial strains from the subset of design candidate microbial strains to thereby create engineered microbial strains;
g. measuring, in an in vitro assay, the phenotypic performance of the engineered microbial strains;
h. selecting a subset of the engineered microbial strains based upon their measured phenotypic performance;
i. adding to the training data set of (a):
i. one or more genetic alteration input variables representing one or more genetic alterations that were introduced into the subset of the engineered microbial strains, and
ii. one or more measured phenotypic performance output variables representing the phenotypic performance measurements of the subset of the engineered microbial strains; and
j. repeating steps (a)-(i) until an engineered microbial strain exhibits a desired level of improved phenotypic performance.
10. The method of claim 9 , wherein the improved phenotypic performance is increased or more efficient production of a product of interest, said product of interest selected from the group consisting of: a small molecule, enzyme, protein, peptide, amino acid, organic acid, synthetic compound, fuel, alcohol, primary extracellular metabolite, secondary extracellular metabolite, intracellular component molecule, and combinations thereof.
11. The method of claim 9 , wherein the improved phenotypic performance is increased or more efficient production of lysine or citric acid.
12. A high-throughput method for engineering a host cell to have improved phenotypic performance, comprising:
a. accessing a training data set containing one or more genetic alteration input variables and one or more measured phenotypic performance output variables,
i. wherein the one or more genetic alteration input variables represent one or more genetic alterations that have been introduced into a host cell through application of one or more libraries, and
ii. wherein the one or more measured phenotypic performance output variables represent one or more phenotypic performance measurements that are associated with the introduced genetic alterations;
b. developing a predictive machine learning model that is populated with the training data set;
c. generating, in silico, a pool of design candidate host cells incorporating the one or more genetic alterations;
d. utilizing the predictive machine learning model to predict the expected phenotypic performance of each member of the pool of design candidate host cells,
i. wherein at least one design candidate host cell comprises a consolidated combination of genetic alterations from among the genetic alterations of step (a), in a genomic sequence, said combination being uncharacterized for improving the phenotypic performance at the time that step (d) is carried out;
ii. wherein the expected phenotypic performance predicted by the machine learning model is based upon the introduced genetic alterations and their associated phenotypic performance measurements of step (a);
e. providing a subset of the design candidate host cells for use in creating engineered host cells; and
f. manufacturing the engineered host cells of step (e).
13. The method of claim 12 , wherein the one or more libraries are selected from the group consisting of: a promoter swap library, a SNP swap library, a start/stop codon library, an optimized sequence library, a terminator swap library, and combinations thereof.
14. The method of claim 12 , wherein the predictive machine learning model incorporates epistatic effects.
15. The method of claim 12 , wherein the predictive machine learning model incorporates at least one of the following: linear regression, kernel ridge regression, logistic regression, neural networks, support vector machines (SVMs), decision trees, hidden Markov models, Bayesian networks, a Gram-Schmidt process, reinforcement-based learning, cluster-based learning, hierarchical clustering, genetic algorithms, or combinations thereof.
16. The method of claim 12 , wherein the predictive machine learning model is supervised, semi-supervised, or unsupervised.
17. The method of claim 12 , wherein the one or more genetic alterations comprise at least one genetic alteration selected from the group consisting of: a single nucleotide polymorphism, nucleotide sequence insertion, nucleotide sequence deletion, and nucleotide sequence replacements.
18. The method of claim 12 , wherein the one or more genetic alterations comprise one or more heterologous promoters from a promoter ladder operably linked to an endogenous target gene.
19. The method of claim 12 , wherein the improved phenotypic performance is increased or more efficient production of a product of interest, said product of interest selected from the group consisting of: a small molecule, enzyme, protein, peptide, amino acid, organic acid, synthetic compound, fuel, alcohol, primary extracellular metabolite, secondary extracellular metabolite, intracellular component molecule, and combinations thereof.