Mutant-Dependent Local Orientational Correlation in Biofilms of Vibrio campbellii Revealed through Digital Processing of Light Microscopy Images.
Maura CesariaMatteo CalcagnilePietro AlifanoRosella CataldoPublished in: International journal of molecular sciences (2023)
Biofilms are key bacterial communities in genetic and adaptive resistance to antibiotics as well as disease control strategies. The mature high-coverage biofilm formations of the Vibrio campbellii strains (wild type BB120 and isogenic derivatives JAF633, KM387, and JMH603) are studied here through the unstraightforward digital processing of morphologically complex images without segmentation or the unrealistic simplifications used to artificially simulate low-density formations. The main results concern the specific mutant- and coverage-dependent short-range orientational correlation as well as the coherent development of biofilm growth pathways over the subdomains of the image. These findings are demonstrated to be unthinkable based only on a visual inspection of the samples or on methods such as Voronoi tessellation or correlation analyses. The presented approach is general, relies on measured rather than simulated low-density formations, and could be employed in the development of a highly efficient screening method for drugs or innovative materials.
Keyphrases
- wild type
- deep learning
- candida albicans
- highly efficient
- biofilm formation
- convolutional neural network
- pseudomonas aeruginosa
- optical coherence tomography
- staphylococcus aureus
- escherichia coli
- affordable care act
- high resolution
- machine learning
- genome wide
- single molecule
- high throughput
- growth factor
- gene expression
- healthcare
- single cell
- high speed
- label free