We also used Approximate Bayesian Computation to fit single-cell genomic sequencing data (from @marcjwills_ @SohrabShah and colleagues) that allows us to infer unobserved properties such as the number of ecDNAs within cells and the rate of chromosomal fusions:

We embed this model into a branching process which allows us to look at the evolutionary consequences of rearrangements.

The model captures the essentials of what happens in G1, S, G2 and M when double strand breaks (DSBs) occur. When there are no cellular checkpoints DSBs are allowed to propagate into S and G2. When these are joined via NHEJ they causing (sometimes massive) rearrangement.

Because an engineered live biotherapeutic (eLBP) requires a balance of growth and killing we explored performance in co-cultures and used Bayesian stats + lotka volterra type models to infer growth rates and interaction parameters, quantifiying production/killing.

We tested different secretion signal peptides and explored how killing depended on them.

The bacteriocin secretion platform uses the SecYEG and GspD systems built within a MoClo framework. As a proof of concept we create Enterocin A and Enterocin B secreting strains that show strong antimicrobial activity against Enterococcus faecalis and Enterococcus faecium

Finally, we extended our experimental system to incorporate biosensing colonies as morphogen sources, demonstrating how a diagnostic device might be constructed.

We demonstrate our approach by constructing a selection of three-input logic gates. These are designed using the Macchiato algorithm to find the minimal set of highpass and bandpass receivers that will encode the given function.

Inspired by electronic design automation, we developed Macchiato, an algorithm for designing optimal spatial circuits computing two-level digital functions. This enhances the capability of our system to complex digital functions without increasing the biological complexity.

We used a liquid handling robot to dispense droplets of culture that form into colonies. We designed spatial patterns of inputs and receiver colonies that are capable of producing the four, non-trivial, two-input logic gates achievable with the highpass and bandpass

Our computers are composed of bacterial colonies which, receive information in the form of diffusible morphogen-like signals. A computation is encoded by the physical locations of the input sources of morphogen and the output receiver colonies.