Search

Multi-factor screenings are commonly used in diverse applications in medicine and bioengineering, including optimizing combination drug treatments and microbiome engineering. Despite the advances in high-throughput technologies, large-scale experiments typically remain prohibitively expensive. Here we introduce a machine learning platform, structure-augmented regression (SAR), that exploits the intrinsic structure of each biological system to learn a high-accuracy model with minimal data requirement. Under different environmental perturbations, each biological system exhibits a unique, structured phenotypic response. This structure can be learned based on limited data and once learned, can constrain subsequent quantitative predictions. We demonstrate that SAR requires significantly fewer data comparing to other existing machine-learning methods to achieve a high prediction accuracy, first on simulated data, then on experimental data of various systems and input dimensions. We then show how a learned structure can guide effective design of new experiments. Our approach has implications for predictive control of biological systems and an integration of machine learning prediction and experimental design. Author summary: Using limited data to predict how biological systems respond to combinations of perturbations is important for better understanding of biological systems and for practical applications. Here we present a new method, structure-augmented regression, that efficiently learns biological response landscapes from sparse measurements. Our method exploits the property that many biological response landscapes have distinct lower-dimensional structures, which can be learned first to assist the subsequent quantitative predictions. We demonstrate that our algorithm outperforms existing algorithms on simulated and experimental data of various biological systems and dimensions. We further exploit the learned structure by suggesting new experiments that refines the learned structure to further improve the prediction accuracy. By integrating machine learning with experimental design, our approach has implications for the predictive control of biological systems. [ABSTRACT FROM AUTHOR]

Fluctuations in the growth rate of a bacterial culture during unbalanced growth are generally considered undesirable in quantitative studies of bacterial physiology. Under well-controlled experimental conditions, however, these fluctuations are not random but instead reflect the interplay between intra-cellular networks underlying bacterial growth and the growth environment. Therefore, these fluctuations could be considered quantitative phenotypes of the bacteria under a specific growth condition. Here, we present a method to identify "phenotypic signatures" by time-frequency analysis of unbalanced growth curves measured with high temporal resolution. The signatures are then applied to differentiate amongst different bacterial strains or the same strain under different growth conditions, and to identify the essential architecture of the gene network underlying the observed growth dynamics. Our method has implications for both basic understanding of bacterial physiology and for the classification of bacterial strains.

Dispersal is necessary for spread into new habitats, but it has also been shown to inhibit spread. Theoretical studies have suggested that the presence of a strong Allee effect may account for these counterintuitive observations. Experimental demonstration of this notion is lacking due to the difficulty in quantitative analysis of such phenomena in a natural setting. We engineered Escherichia coli to exhibit a strong Allee effect and examined how the Allee effect would affect the spread of the engineered bacteria. We showed that the Allee effect led to a biphasic dependence of bacterial spread on the dispersal rate: spread is promoted for intermediate dispersal rates but inhibited at low or high dispersal rates. The shape of this dependence is contingent upon the initial density of the source population. Moreover, the Allee effect led to a tradeoff between effectiveness of population spread and survival: increasing the number of target patches during dispersal allows more effective spread, but it simultaneously increases the risk of failing to invade or of going extinct. We also observed that total population growth is transiently maximized at an intermediate number of target patches. Finally, we demonstrate that fluctuations in cell growth may contribute to the paradoxical relationship between dispersal and spread. Our results provide direct experimental evidence that the Allee effect can explain the apparently paradoxical effects of dispersal on spread and have implications for guiding the spread of cooperative organisms.