Data-Driven Strain Design Using Aggregated Adaptive Laboratory Evolution Mutational Data

Abstract

Microbes are being engineered for an increasingly large and diverse set of applications. However, the designing of microbial genomes remains challenging due to the general complexity of biological system. Adaptive Laboratory Evolution (ALE) leverages nature's problem-solving processes to generate optimized genotypes currently inaccessible to rational methods. The large amount of public ALE data now represents a new opportunity for data-driven strain design. This study presents a novel and first of its kind meta-analysis workflow to derive data-driven strain designs from aggregate ALE mutational data using rich mutation annotations, statistical and structural biology methods. The mutational dataset consolidated and utilized in this study contained 63 Escherichia coli K-12 MG1655 based ALE experiments, described by 93 unique environmental conditions, 357 independent evolutions, and 13,957 observed mutations. High-level trends across the entire dataset were established and revealed that ALE-derived strain designs will largely be gene-centric, as opposed to non-coding, and a relatively small number of variants (approx. 4) can significantly alter cellular states and provide benefits which range from an increase in fitness to a complete necessity for survival. Three novel experimentally validated designs relevant to metabolic engineering applications are presented as use cases for the workflow. Specifically, these designs increased growth rates with glycerol as a carbon source through a point mutation to glpK and a truncation to cyaA or increased tolerance to toxic levels of isobutyric acid through a pykF truncation. These results demonstrate how strain designs can be extracted from aggregated ALE data to enhance strain design efforts.

Meta-analysis Driven Strain Design for Mitigating Oxidative Stresses Important in Biomanufacturing

“…All of the ALE mutations were implemented into the genome of HMP3071 (background strain of HMP3427) using one of the following methods. SDT392 and SDT393 were generated by CRISPR/MAD7 as described previously . SDT711, SDT712, and SDT713 were generated through single target TM-MAGE where repair oligos contain only one editing target.…”

“…Experimental evidence further emphasizes this potential by demonstrating that different substitutions to a single gene can induce a phenotype across a range of intensities . ALEdb (), a publicly accessible database of aggregated ALE mutations, has the potential to provide data that could lead to a more comprehensive understanding of gain-of-function mutations and result in a more effective strain design solution. , …”

“… 28 ALEdb ( ), a publicly accessible database of aggregated ALE mutations, 29 has the potential to provide data that could lead to a more comprehensive understanding of gain-of-function mutations and result in a more effective strain design solution. 30 , 31 …”

Data-Driven Strain Design Towards Mitigating Biomanufacturing Stresses

“…All the ALE mutations were implemented into the genome of HMP3271 (background strain of HMP3427) using one of the following methods. SDT392 and SDT393 were generated by CRISPR/MAD7 as described previously (16). SDT711, SDT712, and SDT713 was generated through single target TM-MAGE (39) where repair oligos contain only one editing target.…”

“…Adaptive Laboratory Evolution (ALE) is an experimental evolution method that has the potential to provide novel solutions to strain design in the form of mutations frequently not observed in natural isolates (13,14). ALEdb (aledb.org), a publicly accessible database of aggregated ALE mutations (15), has the potential to provide data that could lead to a more comprehensive understanding of gain-of-function mutations and result in a more effective strain design solution (16,17). Additional resources that are interoperable with ALEdb data are available for describing the consequences of ALE mutations, ultimately providing insights on which ALE mutations may be most effective for a desired phenotype.…”