Root traits of European Vicia faba cultivars—Using machine learning to explore adaptations to agroclimatic conditions. (16th October 2017)

• Record Type: Journal Article
• Authors: Zhao, Jiangsan; Sykacek, Peter; Bodner, Gernot; Rewald, Boris
• Is Part Of: Plant, cell and environment. Volume 41:Number 9(2018)
• Page Start: 1984
• Page End: 1996
• Publication Date: 2017-10-16
• Subjects: breeding -- faba bean (Vicia faba L.) -- group classification -- kernel spectral clustering -- k‐nearest neighbour -- phenotyping -- random forest -- root traits selection -- supervised learning -- unsupervised learning; Plant physiology -- Periodicals; Plant cells and tissues -- Periodicals; Plant communities -- Periodicals; 581.105
• DOI: 10.1111/pce.13062
• Languages: English
• ISSNs: 0140-7791

Abstract

Abstract: Faba bean ( Vicia faba L.) is an important source of protein, but breeding for increased yield stability and stress tolerance is hampered by the scarcity of phenotyping information. Because comparisons of cultivars adapted to different agroclimatic zones improve our understanding of stress tolerance mechanisms, the root architecture and morphology of 16 European faba bean cultivars were studied at maturity. Different machine learning (ML) approaches were tested in their usefulness to analyse trait variations between cultivars. A supervised, that is, hypothesis‐driven, ML approach revealed that cultivars from Portugal feature greater and coarser but less frequent lateral roots at the top of the taproot, potentially enhancing water uptake from deeper soil horizons. Unsupervised clustering revealed that trait differences between northern and southern cultivars are not predominant but that two cultivar groups, independently from major and minor types, differ largely in overall root system size. Methodological guidelines on how to use powerful ML methods such as random forest models for enhancing the phenotypical exploration of plants are given. Abstract : In the Manuscript (MS), we are able both to demonstrate the correct and novel use of different machine learning methodologies on phenotypical data sets and to unravel the adaptation of faba bean root architecture to Northern and Southern European growing conditions. We, for example, demonstrate that an unsupervised machine learning approach is not able to identify the same traits as a supervised approach but that clusters are dominated by less subtle differences in traits.