Search for: All records

Abstract Synthetic biologists seek to engineer intelligent living systems capable of decision-making, communication, and memory. Separate technologies exist for each tenet of intelligence; however, the unification of all three properties in a living system has not been achieved. Here, we engineer completely intelligentEscherichia colistrains that harbor six orthogonal and inducible genome-integrated recombinases, forming Molecularly Encoded Memory via an Orthogonal Recombinase arraY (MEMORY). MEMORY chassis cells facilitate intelligence via the discrete multi-input regulation of recombinase functions enabling inheritable DNA inversions, deletions, and genomic insertions. MEMORY cells can achieve programmable and permanent gain (or loss) of functions extrachromosomally or from a specific genomic locus, without the loss or modification of the MEMORY platform – enabling the sequential programming and reprogramming of DNA circuits within the cell. We demonstrate all three tenets of intelligence via a probiotic (Nissle 1917) MEMORY strain capable of information exchange with the gastrointestinal commensalBacteroides thetaiotaomicron. 

Abstract Here we present a technology to facilitate synthetic memory in a living system via repurposing Transcriptional Programming (i.e., our decision-making technology) parts, to regulate (intercept) recombinase function post-translation. We show that interception synthetic memory can facilitate programmable loss-of-function via site-specific deletion, programmable gain-of-function by way of site-specific inversion, and synthetic memory operations with nested Boolean logical operations. We can expand interception synthetic memory capacity more than 5-fold for a single recombinase, with reconfiguration specificity for multiple sites in parallel. Interception synthetic memory is ~10-times faster than previous generations of recombinase-based memory. We posit that the faster recombination speed of our next-generation memory technology is due to the post-translational regulation of recombinase function. This iteration of synthetic memory is complementary to decision-making via Transcriptional Programming – thus can be used to develop intelligent synthetic biological systems for myriad applications. 

Abstract Bacteroidesspecies are prominent members of the human gut microbiota. The prevalence and stability ofBacteroidesin humans make them ideal candidates to engineer as programmable living therapeutics. Here we report a biotic decision-making technology in a community ofBacteroides(consortium transcriptional programming) with genetic circuit compression. Circuit compression requires systematic pairing of engineered transcription factors with cognate regulatable promoters. In turn, we demonstrate the compression workflow by designing, building, and testing all fundamental two-input logic gates dependent on the inputs isopropyl-β-D-1-thiogalactopyranoside and D-ribose. We then deploy complete sets of logical operations in five human donorBacteroides, with which we demonstrate sequential gain-of-function control in co-culture. Finally, we couple transcriptional programs with CRISPR interference to achieve loss-of-function regulation of endogenous genes—demonstrating complex control over community composition in co-culture. This work provides a powerful toolkit to program gene expression inBacteroidesfor the development of bespoke therapeutic bacteria. 

Summary Expert based ensemble learning algorithms often serve as online learning algorithms for an unknown, possibly time‐varying, probability distribution. Their simplicity allows flexibility in design choices, leading to variations that balance adaptiveness and consistency. This article provides an analytical framework to quantify the adaptiveness and consistency of expert based ensemble learning algorithms. With properly selected states, the algorithms are modeled as a Markov chains. Then quantitative metrics of adaptiveness and consistency can be calculated through mathematical formulas, other than relying on numerical simulations. Results are derived for several popular ensemble learning algorithms. Success of the method has also been demonstrated in both simulation and experimental results. 

Abstract Traditionally engineered genetic circuits have almost exclusively used naturally occurring transcriptional repressors. Recently, non-natural transcription factors (repressors) have been engineered and employed in synthetic biology with great success. However, transcriptional anti-repressors have largely been absent with regard to the regulation of genes in engineered genetic circuits. Here, we present a workflow for engineering systems of non-natural anti-repressors. In this study, we create 41 inducible anti-repressors. This collection of transcription factors respond to two distinct ligands, fructose (anti-FruR) or D-ribose (anti-RbsR); and were complemented by 14 additional engineered anti-repressors that respond to the ligand isopropyl β-d-1-thiogalactopyranoside (anti-LacI). In turn, we use this collection of anti-repressors and complementary genetic architectures to confer logical control over gene expression. Here, we achieved all NOT oriented logical controls (i.e., NOT, NOR, NAND, and XNOR). The engineered transcription factors and corresponding series, parallel, and series-parallel genetic architectures represent a nascent anti-repressor based transcriptional programming structure. 

Traditional therapeutics aim to diagnose, treat, and cure diseases through various synthetic and natural approaches. The emerging field of engineered living therapeutics (ELTs) genetically functionalizes living cells to alter the paradigm of designed solutions. In this review, we focus on ELTs derived from microbial cell scaffolds. We propose three synergistic modalities for the rational design of ELTs: first, use of regulatory operations to regulate genetic expression; second, integration of alternative biosensing inputs for directed application; third, choice of microbial chassis to deliver solutions. We highlight the challenges and future opportunities within each group and conclude by providing a prospective outlook for ELTs. 

Here we present a technology to facilitate synthetic memory in a living system via repurposing Transcriptional Programming (i.e., our decision-making technology) parts, to regulate (intercept) recombinase function post-translation. We show that interception synthetic memory can facilitate programmable loss-of-function via site-specific deletion, programmable gain-of-function by way of site-specific inversion, and synthetic memory operations with nested Boolean logical operations. We can expand interception synthetic memory capacity more than 5-fold for a single recombinase, with reconfiguration specificity for multiple sites in parallel. Interception synthetic memory is ~10-times faster than previous generations of recombinase-based memory. We posit that the faster recombination speed of our next-generation memory technology is due to the post-translational regulation of recombinase function. This iteration of synthetic memory is complementary to decision-making via Transcriptional Programming – thus can be used to develop intelligent synthetic biological systems for myriad applications.