Answers are grounded in this paper and cite their sources.
Estimating nematode reproduction is crucial in a wide range of research applications, including breeding for host resistance. In root-knot nematodes (RKN), Meloidogyne spp., this is done by extracting eggs from infected roots, diluting the extracted sample, and enumerating by eye under a microscope; a process that is time-intensive, laborious, and requires special training. We developed and tested an automated counting software, NemaQuant, to improve the speed and accuracy of counting. We extracted Meloidogyne incognita and M. enterolobii eggs from 711 plant samples across resistant and susceptible genotypes of sweetpotato, pepper, and watermelon. Four independent aliquots were taken from each sample, serially diluted on a 96-well optical-bottom plate, and imaged using a Keyence automated inverted microscope. From images, NemaQuant was developed using YOLO11 architecture for object detection, trained on 600 images of RKN eggs annotated by nematologists, to count eggs and aggregate counts into a CSV. Across fecundity metrics, total eggs in a sample, eggs per gram of root, and reproduction factor; R2 between human counts and NemaQuant counts ranged from 0.89 to 0.97. We validated the ability of NemaQuant counts to be used for determining resistant genotypes by obtaining the same 44 resistant genotypes as human counts across our four host types. NemaQuant results correlate strongly to traditional methods, take a fraction of the time, and work well in heavy debris ridden backgrounds. We have replaced human RKN counting in our lab with this method, and hope that this study serves as a template for others to do the same.
No takes yet. Share an insight, caveat, or question.
Wram et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: