Distributed Classifier Based on Genetically Engineered Bacterial Cell Cultures

Abstract

We describe a conceptual design of a distributed classifier formed by a population of genetically engineered microbial cells. The central idea is to create a complex classifier from a population of weak or simple classifiers. We create a master population of cells with randomized synthetic biosensor circuits that have a broad range of sensitivities toward chemical signals of interest that form the input vectors subject to classification. The randomized sensitivities are achieved by constructing a library of synthetic gene circuits with randomized control sequences (e.g., ribosome-binding sites) in the front element. The training procedure consists in reshaping of the master population in such a way that it collectively responds to the “positive” patterns of input signals by producing above-threshold output (e.g., fluorescent signal), and below-threshold output in case of the “negative” patterns. The population reshaping is achieved by presenting sequential examples and pruning the population using either graded selection/counterselection or by fluorescence-activated cell sorting (FACS). We demonstrate the feasibility of experimental implementation of such system computationally using a realistic model of the synthetic sensing gene circuits.

Synthetic biology routes to bio-artificial intelligence

“…Thus, the resulting response function of the entire two-promoter circuit to the concentration of signalling molecule is bell shaped for the relevant values of the input signal. ( B ) In the case of two input ranges, X 1 and X 2 , the sensor/output modules feed into an AND gate which sums the output signals as either the presence or absence of GFP expression [30,33]. Adapted with permission from Dydovik et al [30] and Kanakov et al [33].…”

“…To meet these challenges, so-called ‘ensemble’ classifiers have been proposed [30,31]. The ensemble concept requires establishment of a heterogeneous population of simple classifier SGNs that encompasses a random distribution of sensitivities to input signals, each responding to only a narrow range of input levels.…”

Designing Distributed Cell Classifier Circuits using a Genetic Algorithm

“…Such a set can consist of rather simple classifiers that still perform better and more robustly than a complex single circuit classifier, since the individual classification results are aggregated which compensates for individual mistakes. A theoretical design of such a distributed classifier based on synthetic gene circuits was presented by Didovyk et al [3]. The classifier is optimized by training a starting population of simple circuits on the available data similarly to machine learning algorithms, i.e., by presenting learning examples and successively removing low-performance circuits.…”

“…Here, we adapt the distributed classifier approach proposed by Didovyk et al [3] to the problem of cell classifier design. We define a Distributed Classifier (DC) as a set of single-circuit classifiers that decide collectively based on a threshold function.…”

Synthetic Gene Circuits Learn to Classify

“…Boosting the performance of a classifier by combining inputs of multiple distinct weak classifiers is a well-known meta-algorithm in machine learning (Freund & Shapire 1999). Theoretically, the outputs from individual cells can be appropriately combined to produce a robust classification output (Didovyk et al 2015). Of course, design and training of such distributed classifiers present additional challenges that will need to be overcome in the future.…”