SymbiQuant: A Machine Learning Object Detection Tool for Polyploid Independent Estimates of Endosymbiont Population Size

Abstract

Quantifying the size of endosymbiont populations is challenging because endosymbionts are typically difficult or impossible to culture and commonly polyploid. Current approaches to estimating endosymbiont population sizes include quantitative PCR (qPCR) targeting endosymbiont genomic DNA and flow-cytometry. While qPCR captures genome copy number data, it does not capture the number of bacterial cells in polyploid endosymbiont populations. In contrast, flow cytometry can capture accurate estimates of whole host-level endosymbiont population size, but it is not readily able to capture data at the level of endosymbiotic host cells. To complement these existing approaches for estimating endosymbiont population size, we designed and implemented an object detection/segmentation tool for counting the number of endosymbiont cells in micrographs of host tissues. The tool, called SymbiQuant, which makes use of recent advances in deep neural networks includes a graphic user interface that allows for human curation of tool output. We trained SymbiQuant for use in the model aphid/Buchnera endosymbiosis and studied Buchnera population dynamics and phenotype over aphid postembryonic development. We show that SymbiQuant returns accurate counts of endosymbionts, and readily captures Buchnera phenotype. By replacing our training data with data composed of annotated microscopy images from other models of endosymbiosis, SymbiQuant has the potential for broad application. Our tool, which is available on GitHub, adds to the repertoire of methods researchers can use to study endosymbiosis at the organismal, genome, and now endosymbiotic host tissue or cell levels.

Vitellogenin Facilitates Associations between the Whitefly and a Bacteriocyte Symbiont

“…In this study, we refer to the relative symbiont abundance or density between various treatments. Both the aphid and Buchnera experience polyploidy, which dramatically changes throughout insect development for both players ( 58 , 59 ). Whitefly bacteriocytes and associated symbionts may also show polyploidy as in the aphid- Buchnera system.…”

Synthetic Peptide Nucleic Acids in Action:In VivoInterference of thegroELGene from Pea Aphid EndosymbiontBuchnera aphidicola

“…To quantify the extent of Buchnera cell deformation from the imaging data, we used the SymbiQuant software. This machine learning-based object detection tool was designed for the identification and enumeration of endosymbionts in host tissue micrographs [23]. The software authors trained SymbiQuant to detect Buchnera cells using microscopic images captured during normal aphid development [23].…”

“…This machine learning-based object detection tool was designed for the identification and enumeration of endosymbionts in host tissue micrographs [23]. The software authors trained SymbiQuant to detect Buchnera cells using microscopic images captured during normal aphid development [23]. Therefore, we surmised that SymbiQuant primarily identifies and counts spherical Buchnera cells, which is the morphology of normal cells, consequently distinguishing these from those exhibiting abnormal morphologies.…”

“…1.52q) [36]. Quantification of the Buchnera count was performed using SymbiQuant software in Google Colab ( https://github.com/EBJames/SymbiQuant), following the authors' instructions [23]. The Buchnera count for each bacteriocyte was normalized by the bacteriocyte area (µm 2 ) measured with ImageJ.…”

A swarm‐optimized microbial colony counter

“…One application of DL in microbiology is the general detection of biological objects, which primarily involves locating a certain set of objects in an image captured using any biological equipment and classifying them into different types. While a general approach for such applications is to draw a bounding box around the object and classify them into different categories, this approach is not suitable when the shape of the object is a crucial factor for accurate detection (James et al, 2022). In such cases, when the shape is essential for object detection, semantic segmentation is the most preferred DL technique, as it can classify even the smallest unit of an image (Alqazzaz et al, 2019; Amin et al, 2019; Shihavuddin et al, 2021).…”