Separating Soil and Microbial Influences on Soybean Vigour: Methodological Advances in Predictive Modelling

Abstract

Soybean (Glycine max) is a globally important crop, valued for its high protein and oil content seeds. Yet, expansion of soybean cultivation has come with considerable environmental costs. Addressing the challenges of food security and ecosystem preservation demands not just efficient crop production, but more sustainable agricultural strategies that harness natural biological interactions. One promising solution lies in leveraging the plant microbiome. Beneficial microbes, such as rhizobia, can enhance crop performance and resilience. However, predicting which microbes will benefit plants under real field conditions remains a central challenge: the effects of a given microbe depend on complex interactions with both abiotic and biotic influences. Recent work by Song et al. (2024) provided a powerful framework for a microbiome-data driven predictive model, using both high-resolution sequencing and phenotyping to model potato vigour from soil microbial community profiles. Inspired by this approach, this thesis lays the groundwork for a similar predictive modelling effort in soybean. To support future modelling, a series of foundational experiments was undertaken: (1) chemical and physical analysis of previously collected soils, (2) development and validation of whole-microbiome transfer methods to separate microbial effects from abiotic soil effects, (3) attempts to chromosomally tag rhizobial strains for precise tracking and (4) tracing of the fate of tagged rhizobia in complex communities, which resulted unsuccessfully. By integrating the results of these experiments, and drawing on the example set by Song et al. (2024), the next phase will seek to identify “microbial collaboromes” and develop a predictive model of soybean vigour, offering a scientific basis for microbiome-informed, sustainable crop management.