Machine Learning-Based Plant Detection Algorithms to Automate Counting Tasks Using 3D Canopy Scans.

This study tested whether machine learning (ML) methods can effectively separate individual plants from complex 3D canopy laser scans as a prerequisite to analyzing particular plant features. For this, we scanned mung bean and chickpea crops with PlantEye (R) laser scanners. Firstly, we segmented the crop canopies from the background in 3D space using the Region Growing Segmentation algorithm. Then, Convolutional Neural Network (CNN) based ML algorithms were fine-tuned for plant counting. Application of the CNN-based (Convolutional Neural Network) processing architecture was possible only after we reduced the dimensionality of the data to 2D. This allowed for the identification of individual plants and their counting with an accuracy of 93.18% and 92.87% for mung bean and chickpea plants, respectively. These steps were connected to the phenotyping pipeline, which can now replace manual counting operations that are inefficient, costly, and error-prone. The use of CNN in this study was innovatively solved with dimensionality reduction, addition of height information as color, and consequent application of a 2D CNN-based approach. We found there to be a wide gap in the use of ML on 3D information. This gap will have to be addressed, especially for more complex plant feature extractions, which we intend to implement through further research.

Using Growth and Transpiration Phenotyping Under Controlled Conditions to Select Water Efficient Banana Genotypes.

Water deficit is one of the world's major constraints in agriculture and will aggravate in the future. Banana (Musa spp.) is an important crop that needs vast amounts of water for optimal production. The International Transit Center of Bioversity International holds the world's biggest collection of banana biodiversity (>1,500 accessions). The long-term aim of this research is to evaluate the potential within this collection for climate smart agricultural usage. Therefore, we developed a phenotyping setup under controlled environmental conditions and we selected 32 representatives of the Musa biodiversity (29 cultivars and 3 wild relatives) for evaluation. The best performing genotypes accumulated six to seven times more biomass than the least performing. Eight genotypes (five ABB, one AAB, and two AAA) invest under osmotic stress significantly more in root growth than in leaf growth. We predict therefore that these genotypes have potential for high productivity under rain fed conditions with a short dry season. To gain more insight in the transpiration physiology, we gravimetrically monitored individual plant transpiration over the diurnal period. All analyzed genotypes showed a marked reduction in transpiration rate in the afternoon. Moreover, the timing of this onset, as well as its impact on total transpiration, was genotype dependent. This phenomenon was more pronounced in 13 genotypes (eight ABB, two AAB, two AA, one BB). Banana is a crop originating from the humid tropics and has developed a strong root pressure to maintain an efficient water and nutrient transport even under saturated relative humidity conditions. Therefore, we hypothesize that the diurnal transpiration decline contributes to a higher water use efficiency without compromising the nutrient transport. Of the eight genotypes that had the best growth under osmotic stress, all analyzed ABB cultivars have a lower maximal transpiration rate, keep this maximal transpiration for a shorter time and therefore consume less water per day. We conclude that lab models are very useful to study the biodiversity and to identify different traits that contribute to a better drought tolerance/avoidance. We encourage researchers investigating other crops to start exploring their collections.

Effect of paleopolyploidy and allopolyploidy on gene expression in banana.

BACKGROUND: Bananas (Musa spp.) are an important crop worldwide. Most modern cultivars resulted from a complex polyploidization history that comprised three whole genome duplications (WGDs) shaping the haploid Musa genome, followed by inter- and intra-specific crosses between Musa acuminata and M. balbisiana (A and B genome, respectively). Unresolved hybridizations finally led to banana diversification into several autotriploid (AAA) and allotriploid cultivars (AAB and ABB). Using transcriptomic data, we investigated the impact of the genome structure on gene expression patterns in roots of 12 different triploid genotypes covering AAA, AAB and ABB subgenome constitutions. RESULTS: We demonstrate that (i) there are different genome structures, (ii) expression patterns go beyond the predicted genomic groups, and (iii) the proportion of the B genome influences the gene expression. The presence of the B genome is associated with a higher expression of genes involved in flavonoid biosynthesis, fatty acid metabolism, amino sugar and nucleotide sugar metabolism and oxidative phosphorylation. There are cultivar-specific chromosome regions with biased B:A gene expression ratios that demonstrate homoeologous exchanges (HE) between A and B sub-genomes. In two cultivars, aneuploidy was detected. We identified 3674 genes with a different expression level between allotriploid and autotriploid with ~ 57% having recently duplicated copies (paralogous). We propose a Paralog Inclusive Expression (PIE) analysis that appears to be suitable for genomes still in a downsizing and fractionation process following whole genome duplications. Our approach allows highlighting the genes with a maximum likelihood to affect the plant phenotype. CONCLUSIONS: This study on banana is a good case to investigate the effects of alloploidy in crops. We conclude that allopolyploidy triggered changes in the genome structure of a crop and it clearly influences the gene.

Homeolog expression analysis in an allotriploid non-model crop via integration of transcriptomics and proteomics.

The fate of doubled genes, from allopolyploid or autopolyploid origin, is controlled at multiple levels, resulting in the modern day cultivars. We studied the root growth of 3 different triploid banana cultivars under control and osmotic stress conditions. The root growth of the allopolyploid ABB cultivar was 42% higher under control and 61% higher under osmotic stress. By integrating transcriptomics and proteomics, we studied the gene expression of all 3 cultivars, resulting in 2,749 identified root proteins. 383 gene loci displayed genotype specific differential expression whereof 252 showed at least one Single Amino Acid Polymorphism (SAAP). In the ABB cultivar, allele expressions supposedly follow a 1/3 and 2/3 pattern for respectively the A and the B allele. Using transcriptome read alignment to assess the homeoallelic contribution we found that 63% of the allele specific genes deviated from this expectation. 32 gene loci even did not express the A allele. The identified ABB allele- specific proteins correlate well with the observed growth phenotype as they are enriched in energy related functions such as ATP metabolic processes, nicotinamide nucleotide metabolic processes, and glycolysis.

Differential root transcriptomics in a polyploid non-model crop: the importance of respiration during osmotic stress.

To explore the transcriptomic global response to osmotic stress in roots, 18 mRNA-seq libraries were generated from three triploid banana genotypes grown under mild osmotic stress (5% PEG) and control conditions. Illumina sequencing produced 568 million high quality reads, of which 70-84% were mapped to the banana diploid reference genome. Using different uni- and multivariate statistics, 92 genes were commonly identified as differentially expressed in the three genotypes. Using our in house workflow to analyze GO enriched and underlying biochemical pathways, we present the general processes affected by mild osmotic stress in the root and focus subsequently on the most significantly overrepresented classes associated with: respiration, glycolysis and fermentation. We hypothesize that in fast growing and oxygen demanding tissues, mild osmotic stress leads to a lower energy level, which induces a metabolic shift towards (i) a higher oxidative respiration, (ii) alternative respiration and (iii) fermentation. To confirm the mRNA-seq results, a subset of twenty up-regulated transcripts were further analysed by RT-qPCR in an independent experiment at three different time points. The identification and annotation of this set of genes provides a valuable resource to understand the importance of energy sensing during mild osmotic stress.

Overexpression of the trehalase gene AtTRE1 leads to increased drought stress tolerance in Arabidopsis and is involved in abscisic acid-induced stomatal closure.

Introduction of microbial trehalose biosynthesis enzymes has been reported to enhance abiotic stress resistance in plants but also resulted in undesirable traits. Here, we present an approach for engineering drought stress tolerance by modifying the endogenous trehalase activity in Arabidopsis (Arabidopsis thaliana). AtTRE1 encodes the Arabidopsis trehalase, the only enzyme known in this species to specifically hydrolyze trehalose into glucose. AtTRE1-overexpressing and Attre1 mutant lines were constructed and tested for their performance in drought stress assays. AtTRE1-overexpressing plants had decreased trehalose levels and recovered better after drought stress, whereas Attre1 mutants had elevated trehalose contents and exhibited a drought-susceptible phenotype. Leaf detachment assays showed that Attre1 mutants lose water faster than wild-type plants, whereas AtTRE1-overexpressing plants have a better water-retaining capacity. In vitro studies revealed that abscisic acid-mediated closure of stomata is impaired in Attre1 lines, whereas the AtTRE1 overexpressors are more sensitive toward abscisic acid-dependent stomatal closure. This observation is further supported by the altered leaf temperatures seen in trehalase-modified plantlets during in vivo drought stress studies. Our results show that overexpression of plant trehalase improves drought stress tolerance in Arabidopsis and that trehalase plays a role in the regulation of stomatal closure in the plant drought stress response.