Pooled genomic library assembly, sequencing and scoring for multi-objective phenotype optimization

Precision engineering of complex behaviors in microbes is limited by our understanding of the contexts of gene expression and ability to manipulate DNA in individual cells at scale. Strain libraries built by iterative design, build and test cycles often sample only a small region of the functional parameter space. Further, the relative ease of generating expression libraries to optimize gene networks on plasmids often is not predictive of system performance when transferred to the chromosome for deployment in complex environments ranging from the bioreactor to the mammalian gut.

We have developed poolGLASS, an engineering platform for pooled Genomic Library Assembly, Sequencing and Scoring coupled with CRISPR interference-mediated retrieval of individual genotypes from the library. We validated poolGLASS by creating a barcoded combinatorial genomic library of over two million ribosome binding site variants of the violacein biosynthetic pathway vioABEDC in E. coli, sampling from over 260,000 combinatorial genotypes. We used emulsion PCR and high-throughput sequencing to genotype the library and quantified variant fitness in parallel by tracking barcode abundance under pathway induction and in resource competition with a violacein-sensitive microbe. Barcode fluctuation analysis revealed genotypes that optimally balance the cost of gene expression against the benefits of antimicrobial production. We isolated top variants from the library by barcode-specific repression of GFP coupled with cell sorting. Through comprehensive sampling of the expression space of a target gene network, poolGLASS enables deep mapping of sequence to function relationships and rapid optimization of cryptic biosynthetic gene clusters and synthetic gene circuits.