An Automated Image Analysis Method for Segmenting Fluorescent Bacteria in Three Dimensions

Single-cell fluorescence
imaging is a powerful technique for studying
inherently heterogeneous biological processes. To correlate a genotype
or phenotype to a specific cell, images containing a population of
cells must first be properly segmented. However, a proper segmentation
with minimal user input becomes challenging when cells are clustered
or overlapping in three dimensions. We introduce a new analysis package,
Seg-3D, for the segmentation of bacterial cells in three-dimensional
(3D) images, based on local thresholding, shape analysis, concavity-based
cluster splitting, and morphology-based 3D reconstruction. The reconstructed
cell volumes allow us to directly quantify the fluorescent signals
from biomolecules of interest within individual cells. We demonstrate
the application of this analysis package in 3D segmentation of individual
bacterial pathogens invading host cells. We believe Seg-3D can be
an efficient and simple program that can be used to analyze a wide
variety of single-cell images, especially for biological systems involving
random 3D orientation and clustering behavior, such as bacterial infection
or colonization.