Search

Improvements in microscopy software and hardware have dramatically increased the pace of image acquisition, making analysis a major bottleneck in generating quantitative, single-cell data. Although tools for segmenting and tracking bacteria within time-lapse images exist, most require human input, are specialized to the experimental set up, or lack accuracy. Here, we introduce DeLTA 2.0, a purely Python workflow that can rapidly and accurately analyze images of single cells on two-dimensional surfaces to quantify gene expression and cell growth. The algorithm uses deep convolutional neural networks to extract single-cell information from time-lapse images, requiring no human input after training. DeLTA 2.0 retains all the functionality of the original version, which was optimized for bacteria growing in the mother machine microfluidic device, but extends results to two-dimensional growth environments. Two-dimensional environments represent an important class of data because they are more straightforward to implement experimentally, they offer the potential for studies using co-cultures of cells, and they can be used to quantify spatial effects and multi-generational phenomena. However, segmentation and tracking are significantly more challenging tasks in two-dimensions due to exponential increases in the number of cells. To showcase this new functionality, we analyze mixed populations of antibiotic resistant and susceptible cells, and also track pole age and growth rate across generations. In addition to the two-dimensional capabilities, we also introduce several major improvements to the code that increase accessibility, including the ability to accept many standard microscopy file formats as inputs and the introduction of a Google Colab notebook so users can try the software without installing the code on their local machine. DeLTA 2.0 is rapid, with run times of less than 10 minutes for complete movies with hundreds of cells, and is highly accurate, with error rates around 1%, making it a powerful tool for analyzing time-lapse microscopy data., Author summary Time-lapse microscopy can generate large image datasets which track single-cell properties like gene expression or growth rate over time. Deep learning tools are very useful for analyzing these data and can identify the location of cells and track their position. In this work, we introduce a new version of our Deep Learning for Time-lapse Analysis (DeLTA) software, which includes the ability to robustly segment and track bacteria that are growing in two dimensions, such as on agarose pads or within microfluidic environments. This capability is essential for experiments where spatial and positional effects are important, such as conditions with microbial co-cultures, cell-to-cell interactions, or spatial patterning. The software also tracks pole age and can be used to analyze replicative aging. These new features join other improvements, such as the ability to work directly with many common microscopy file formats. DeLTA 2.0 can reliably track hundreds of cells with low error rates, making it an ideal tool for high throughput analysis of microscopy data.

We describe a spatial Moran model that captures mechanical interactions and directional growth in spatially extended populations. The model is analytically tractable and completely solvable under a mean-field approximation and can elucidate the mechanisms that drive the formation of population-level patterns. As an example we model a population of E. coli growing in a rectangular microfluidic trap. We show that spatial patterns can arise as a result of a tug-of-war between boundary effects and growth rate modulations due to cell–cell interactions: Cells align parallel to the long side of the trap when boundary effects dominate. However, when cell–cell interactions exceed a critical value, cells align orthogonally to the trap’s long side. This modeling approach and analysis can be extended to directionally-growing cells in a variety of domains to provide insight into how local and global interactions shape collective behavior.

Our work targets automated analysis to quantify the growth dynamics of a population of bacilliform bacteria. We propose an innovative approach to frame-sequence tracking of deformable-cell motion by the automated minimization of a new, specific cost functional. This minimization is implemented by dedicated Boltzmann machines (stochastic recurrent neural networks). Automated detection of cell divisions is handled similarly by successive minimizations of two cost functions, alternating the identification of children pairs and parent identification. We validate the proposed automatic cell tracking algorithm using (i) recordings of simulated cell colonies that closely mimic the growth dynamics of E. coli in microfluidic traps and (ii) real data. On a batch of 1100 simulated image frames, cell registration accuracies per frame ranged from 94.5% to 100%, with a high average. Our initial tests using experimental image sequences (i.e., real data) of E. coli colonies also yield convincing results, with a registration accuracy ranging from 90% to 100%., 35 pages, 8 figures, 7 tables

Synthetic microbial consortia consist of two or more engineered strains that grow together and share the same resources. When intercellular signaling pathways are included in the engineered strains, close proximity of the microbes can generate complex dynamic behaviors that are difficult to obtain using a single strain. However, when a consortium is not cultured in a well-mixed environment the constituent strains passively compete for space as they grow and divide, complicating cell-cell signaling. Here, we explore the temporal dynamics of the spatial distribution of consortia cocultured in microfluidic devices. To do this, we grew two different strains of Escherichia coli in microfluidic devices with cell-trapping regions (traps) of several different designs. We found that the size of the traps is a critical determinant of spatiotemporal dynamics. In small traps, cells can easily signal one another, but the relative proportion of each strain within the trap can fluctuate wildly. In large traps, the relative ratio of strains is stabilized, but intercellular signaling can be hindered by distances between cells. This presents a trade-off between the trap size and the effectiveness of intercellular signaling, which can be mitigated by increasing the initial seeding of cells in larger traps. We also built a mathematical model, which suggests that increasing the number of seed cells can also increase the strain ratio variability due to an increased number of strain interfaces in the trap. These results help elucidate the complex behaviors of synthetic microbial consortia in microfluidic traps and provide a means of analysis to help remedy the spatial heterogeneity inherent to different trap types.

Bacteriophytochrome photoreceptors (BphP) are knotted proteins that have been developed as near-infrared fluorescent protein (iRFP) reporters of gene expression. To explore how rearrangements in the peptides that interlace into the knot within the BphP photosensory core affect folding, we subjected iRFPs to random circular permutation using an improved transposase mutagenesis strategy and screened for variants that fluoresce. We identified 27 circularly permuted iRFPs that display biliverdin-dependent fluorescence in Escherichia coli. The variants with the brightest whole cell fluorescence initiated translation at residues near the domain linker and knot tails, although fluorescent variants that initiated translation within the PAS and GAF domains were discovered. Circularly permuted iRFPs retained sufficient cofactor affinity to fluoresce in tissue culture without the addition of biliverdin, and one variant displayed enhanced fluorescence when expressed in bacteria and tissue culture. This variant displayed a quantum yield similar to that of iRFPs but exhibited increased resistance to chemical denaturation, suggesting that the observed increase in the magnitude of the signal arose from more efficient protein maturation. These results show how the contact order of a knotted BphP can be altered without disrupting chromophore binding and fluorescence, an important step toward the creation of near-infrared biosensors with expanded chemical sensing functions for in vivo imaging.

We introduce a tractable stochastic spatial Moran model to explain experimentally-observed patterns of rod-shaped bacteria growing in rectangular microfluidic traps. Our model shows that spatial patterns can arise as a result of a tug-of-war between boundary effects and modulations of growth rate due to cell-cell interactions. Cells alignparallelto the long side of the trap when boundary effects dominate. However, when the magnitude of cell-cell interactions exceeds a critical value, cells align orthogonally to the trap’s long side. Our model is analytically tractable, and completely solvable under a mean-field approximation. This allows us to elucidate the mechanisms that govern the formation of population-level patterns. The model can be easily extended to examine various types of interactions that can shape the collective behavior in bacterial populations.

Synthetic microbial consortia have an advantage over isogenic synthetic microbes because they can apportion biochemical and regulatory tasks among the strains. However, it is difficult to coordinate gene expression in spatially extended consortia because the range of signaling molecules is limited by diffusion. Here, we show that spatio-temporal coordination of gene expression can be achieved even when the spatial extent of the consortium is much greater than the diffusion distance of the signaling molecules. To do this, we examined the dynamics of a two-strain synthetic microbial consortium that generates coherent oscillations in small colonies. In large colonies, we find that temporally coordinated oscillations across the population depend on the presence of an intrinsic positive feedback loop that amplifies and propagates intercellular signals. These results demonstrate that synthetic multicellular systems can be engineered to exhibit coordinated gene expression using only transient, short-range coupling among constituent cells.

Until recently, evolutionary questions surrounding the nature of the genetic code have been mostly limited to the realm of conjecture, modeling, and simulation due to the difficulty of altering this fundamental property of living organisms. Concerted genome and protein engineering efforts now make it possible to experimentally study the impact of alternative genetic codes on the evolution of biological systems. We explored how Escherichia coli strains that incorporate a 21st nonstandard amino acid (nsAA) at the recoded amber (TAG) stop codon evolve resistance to the antibiotic rifampicin. Resistance to rifampicin arises from chromosomal mutations in the β subunit of RNA polymerase (RpoB). We found that a variety of mutations that lead to substitutions of nsAAs in the essential RpoB protein confer robust rifampicin resistance. We interpret these results in a framework in which an expanded code can increase evolvability in two distinct ways: by adding a new letter with unique chemical properties to the protein alphabet and by altering the mutational connectivity of amber-adjacent codons by converting a lethal nonsense mutation into a missense mutation. Finally, we consider the implications of these results for the evolution of alternative genetic codes. In our experiments, reliance on a mutation to a reassigned codon for a vital trait is not required for the long-term maintenance of an expanded genetic code and may even destabilize incorporation of an nsAA, a result that is consistent with the codon capture model of genetic code evolution.

The Registry of Standard Biological Parts only accepts genetic parts compatible with the RFC 10 BioBrick format. This combined assembly and submission standard requires that four unique restriction enzyme sites must not occur in the DNA sequence encoding a part. We present evidence that this requirement places a nontrivial burden on iGEM teams developing large and novel parts. We further argue that the emergence of inexpensive DNA synthesis and versatile assembly methods reduces the utility of coupling submission and assembly standards and propose a submission standard that is compatible with current quality control strategies while nearly eliminating sequence constraints on submitted parts.