Investigating Carboxysome Morphology Dynamics with a Rotationally Invariant Variational Autoencoder

Abstract

Carboxysomes are a class of bacterial microcompartments that form proteinaceous organelles within the cytoplasm of cyanobacteria and play a central role in photosynthetic metabolism by defining a cellular microenvironment permissive to CO2 fixation. Critical aspects of the assembly of the carboxysomes remain relatively unknown, especially with regard to the dynamics of this microcompartment. Progress in understanding carboxysome dynamics is impeded in part because analysis of the subtle changes in carboxysome morphology with microscopy remains a low-throughput and subjective process. Here we use deep learning techniques, specifically a Rotationally Invariant Variational Autoencoder (rVAE), to analyze fluorescence microscopy images of cyanobacteria bearing a carboxysome reporter and quantitatively evaluate how carboxysome shell remodelling impacts subtle trends in the morphology of the microcompartment over time. Toward this goal, we use a recently developed tool to control endogenous protein levels, including carboxysomal components, in the model cyanobacterium Synechococcous elongatus PCC 7942. By utilization of this system, proteins that compose the carboxysome can be tuned in real time as a method to examine carboxysome dynamics. We find that rVAEs are able to assist in the quantitative evaluation of changes in carboxysome numbers, shape, and size over time. We propose that rVAEs may be a useful tool to accelerate the analysis of carboxysome assembly and dynamics in response to genetic or environmental perturbation and may be more generally useful to probe regulatory processes involving a broader array of bacterial microcompartments.

In Vitro Analysis of Bacterial Microcompartments and Shell Protein Superstructures by Confocal Microscopy

“…Furthermore, bacteria that encode multiple different BMCs heavily regulate them to prevent their intermixing ( 46 ). Recent data also demonstrate that preformed carboxysomes can dynamically remodel their shell in vivo ( 47 ). It may be the case that BMCs are not strictly rigid in order to be more responsive to environmental changes, perhaps so that they can be rapidly assembled and disassembled based on the cell’s metabolic needs.…”

Combining machine learning with high-content imaging to infer ciprofloxacin susceptibility in clinical isolates of Salmonella Typhimurium

“…While we created an HCI analysis pipeline suitable for single strain bacterial cultures, it likely needs further re nement to optimally segment bacterial cells, and our use of a proprietary analysis software may have limited the parameter exibility. However, similar initial image segmentation and analysis can be performed using state-ofthe-art deep learning architecture, including U-net 34 and Autoencoder 35 . An improved image analysis algorithm specialized for bacterial imaging data may provide improved differentiation between resistant and susceptible organisms at the singlecell level.…”

Combining variational autoencoders and physical bias for improved microscopy data analysis ∗

“…For systems with a well-defined crystallographic lattice, methods based on linear decompositions and variational autoencoders (VAEs) [30] were shown to be highly efficient. The development of the rotationally invariant autoencoders extended these approaches to general orientation and discovery of chemical transformation pathways in disordered systems [31][32][33][34]. Also, such rotationally invariant autoencoders have been tuned with Bayesian optimization to maximize uncovering of the features from complex microscopic data [35].…”