Deep learning-based quantification of arbuscular mycorrhizal fungi in plant roots

Abstract

Soil fungi establish mutualistic interactions with the roots of most vascular land plants. Arbuscular mycorrhizal (AM) fungi are among the most extensively characterised mycobionts to date. Current approaches to quantifying the extent of root colonisation and the relative abundance of intraradical hyphal structures in mutant roots rely on staining and human scoring involving simple, yet repetitive tasks prone to variations between experimenters. We developed the software AMFinder which allows for automatic computer vision-based identification and quantification of AM fungal colonisation and intraradical hyphal structures on ink-stained root images using convolutional neural networks. AMFinder delivered high-confidence predictions on image datasets of colonised roots of Medicago truncatula, Lotus japonicus, Oryza sativa and Nicotiana benthamiana obtained via flatbed scanning or digital microscopy enabling reproducible and transparent data analysis. A streamlined protocol for sample preparation and imaging allowed us to quantify dynamic increases in colonisation in whole root systems over time. AMFinder adapts to a wide array of experimental conditions. It enables accurate, reproducible analyses of plant root systems and will support better documentation of AM fungal colonisation analyses. AMFinder can be accessed here: https://github.com/SchornacklabSLCU/amfinder.git

Mycorrhization in trees: ecology, physiology, emerging technologies and beyond

“…It has been successfully used to characterize AMF‐colonized roots of Nicotiana benthamiana (Benth), Medicago truncatula , Lotus japonicas , and Oryza sativa (Evangelisti et al . 2021). This AMFinder can be readily extended to a range of fungal strains and host species, fungal staining protocols, and set up conditions.…”

“…The Automatic Mycorrhiza Finder (AMFinder) software supports efficient identification of AMF networks and intra-radical hyphal structures by means of computer-vison technology. It has been successfully used to characterize AMF-colonized roots of Nicotiana benthamiana (Benth), Medicago truncatula, Lotus japonicas, and Oryza sativa (Evangelisti et al 2021). This AMFinder can be readily extended to a range of fungal strains and host species, fungal staining protocols, and set up conditions.…”

Antibiotic Combination Therapy: A Strategy to Overcome Bacterial Resistance to Aminoglycoside Antibiotics

“…The most common gram-negative member of the AME genes they carry is aac (6′) -Ib (Ramirez and Tolmasky, 2010;Shaul et al, 2011;Herzog et al, 2012). N-acetyltransferase AAC (3)-I was established as a selectable marker for gentamicin-based oomycete transformation, and it was found that N-acetyltransferase AAC (3)-I was responsible for gentamicin resistance in phytophthora palmivora and phytophthora infestans (Evangelisti et al, 2019).…”

A snapshot of the root phenotyping landscape in 2021

“…Amplicon sequencing of marker genes has become a method of choice for characterising soil and plant-associated microbiota (Schöler et al ., 2017; Nannipieri et al ., 2019). The throughput of root image analysis pipelines has also greatly benefited from the development of tools relying on deep learning for root tip detection (Pound et al ., 2017) or the quantification of plant-associated fungi (Evangelisti et al ., 2021). Moreover, the complementarity that can exist between free, open access and high-performance software packages, as is the case with RootPainter (Smith et al ., 2020) and RhizoVision Explorer (Seethepalli et al ., 2021), holds great promise for eliminating the image analysis bottleneck that root researchers often face in their daily work.…”