AUTHOR=Devarajan Arun Kumar , Truu Marika , Gopalasubramaniam Sabarinathan Kuttalingam , Muthukrishanan Gomathy , Truu Jaak 

TITLE=Application of data integration for rice bacterial strain selection by combining their osmotic stress response and plant growth-promoting traits

JOURNAL=Frontiers in Microbiology

VOLUME=Volume 13 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/microbiology/articles/10.3389/fmicb.2022.1058772

DOI=10.3389/fmicb.2022.1058772

ISSN=1664-302X

ABSTRACT=The application of plant-beneficial bacteria in agriculture to improve crop yield and alleviate stress caused by environmental conditions, pests, and pathogens is gaining popularity. However, before using those bacterial strains in plant experiments, their environmental stress response and potential for plant health improvement should be examined by multiple direct and indirect beneficial traits. This study aimed to explore the applicability of data integration methods for selecting the osmotic stress tolerant plant-growth promoting (PGP) bacterial strains isolated from the rice phyllosphere. For this purpose, three unsupervised machine learning techniques were implemented: principal component analysis (PCA) on concatenated data, multiple co-inertia analysis (MCIA), and multiple kernel learning (MKL). The datasets included in the analyses were measurements of direct and indirect microbial PGP activities and osmotic stress response of eight bacterial strains previously isolated from the leaf surface of drought-tolerant rice cultivars. Our results indicate that data integration methods complement the single table data analysis approach and improve the selection process of plant growth-promoting microbial strains.