BackgroundPhenotyping is a critical component of plant research. Accurate and precise trait collection, when integrated with genetic tools, can greatly accelerate the rate of genetic gain in crop improvement. However, efficient and automatic phenotyping of traits across large populations is a challenge; which is further exacerbated by the necessity of sampling multiple environments and growing replicated trials. A promising approach is to leverage current advances in imaging technology, data analytics and machine learning to enable automated and fast phenotyping and subsequent decision support. In this context, the workflow for phenotyping (image capture → data storage and curation → trait extraction → machine learning/classification → models/apps for decision support) has to be carefully designed and efficiently executed to minimize resource usage and maximize utility. We illustrate such an end-to-end phenotyping workflow for the case of plant stress severity phenotyping in soybean, with a specific focus on the rapid and automatic assessment of iron deficiency chlorosis (IDC) severity on thousands of field plots. We showcase this analytics framework by extracting IDC features from a set of ~4500 unique canopies representing a diverse germplasm base that have different levels of IDC, and subsequently training a variety of classification models to predict plant stress severity. The best classifier is then deployed as a smartphone app for rapid and real time severity rating in the field.ResultsWe investigated 10 different classification approaches, with the best classifier being a hierarchical classifier with a mean per-class accuracy of ~96%. We construct a phenotypically meaningful ‘population canopy graph’, connecting the automatically extracted canopy trait features with plant stress severity rating. We incorporated this image capture → image processing → classification workflow into a smartphone app that enables automated real-time evaluation of IDC scores using digital images of the canopy.ConclusionWe expect this high-throughput framework to help increase the rate of genetic gain by providing a robust extendable framework for other abiotic and biotic stresses. We further envision this workflow embedded onto a high throughput phenotyping ground vehicle and unmanned aerial system that will allow real-time, automated stress trait detection and quantification for plant research, breeding and stress scouting applications.Electronic supplementary materialThe online version of this article (doi:10.1186/s13007-017-0173-7) contains supplementary material, which is available to authorized users.

No abstract

Common bean (Phaseolus vulgaris L.), one of the most important grain legume crops for direct human consumption, faces many challenges as a crop. Domesticated from wild relatives that inhabit a relatively narrow ecological niche, common bean faces a wide range of biotic and abiotic constraints within its diverse agroecological settings. Biotic stresses impacting common bean include numerous bacterial, fungal, and viral diseases and various insect and nematode pests, and abiotic stresses include drought, heat, cold, and soil nutrient deficiencies or toxicities. Breeding is often local, focusing on improvements in responses to biotic and abiotic stresses that are particular challenges in certain locations and needing to respond to conditions such as day-length regimes. This review describes the major breeding objectives for common bean, followed by a description of major genetic and genomic resources, and an overview of current and prospective marker-assisted methods in common bean breeding. Improvements over traditional breeding methods in CB can result from the use of different approaches. Several important germplasm collections have been densely genotyped, and relatively inexpensive SNP genotyping platforms enable implementation of genomic selection and related markerassisted breeding approaches. Also important are sociological insights related to demand-led breeding, which considers local value chains, from farmers to traders to retailers and consumers.

Traditional evaluation of crop biotic and abiotic stresses are time-consuming and labor-intensive limiting the ability to dissect the genetic basis of quantitative traits. A machine learning (ML)-enabled image-phenotyping pipeline for the genetic studies of abiotic stress iron deficiency chlorosis (IDC) of soybean is reported. IDC classification and severity for an association panel of 461 diverse plant-introduction accessions was evaluated using an end-to-end phenotyping workflow. The workflow consisted of a multi-stage procedure including: (1) optimized protocols for consistent image capture across plant canopies, (2) canopy identification and registration from cluttered backgrounds, (3) extraction of domain expert informed features from the processed images to accurately represent IDC expression, and (4) supervised ML-based classifiers that linked the automatically extracted features with expert-rating equivalent IDC scores. ML-generated phenotypic data were subsequently utilized for the genome-wide association study and genomic prediction. The results illustrate the reliability and advantage of ML-enabled image-phenotyping pipeline by identifying previously reported locus and a novel locus harboring a gene homolog involved in iron acquisition. This study demonstrates a promising path for integrating the phenotyping pipeline into genomic prediction, and provides a systematic framework enabling robust and quicker phenotyping through ground-based systems.

Background Breeding programs benefit from information about marker-trait associations for many traits, whether the goal is to place those traits under active selection or to maintain them through background selection. Association studies are also important for identifying accessions bearing potentially useful alleles by characterizing marker-trait associations and allelic states across germplasm collections. This study reports the results of a genome-wide association study and evaluation of epistatic interactions for four agronomic and seed-related traits in soybean. Results Using 419 diverse soybean accessions, together with genotyping data from the SoySNP50K Illumina Infinium BeadChip, we identified marker-trait associations for internode number (IN), plant height (PH), seed weight (SW), and seed yield per plant (SYP). We conducted a genome-wide epistatic study (GWES), identifying candidate genes that show evidence of SNP-SNP interactions. Although these candidate genes will require further experimental validation, several appear to be involved in developmental processes related to the respective traits. For IN and PH, these include the Dt1 determinacy locus (a soybean meristematic transcription factor), as well as a pectinesterase gene and a squamosa promoter binding gene that in other plants are involved in cell elongation and the vegetative-to-reproductive transition, respectively. For SW, candidate genes include an ortholog of the AP2 gene, which in other species is involved in maintaining seed size, embryo size, seed weight and seed yield. Another SW candidate gene is a histidine phosphotransfer protein - orthologs of which are involved in cytokinin-mediated seed weight regulating pathways. The SYP association loci overlap with regions reported in previous QTL studies to be involved in seed yield. Conclusions This study further confirms the utility of GWAS and GWES approaches for identifying marker-trait associations and interactions within a diverse germplasm collection. Electronic supplementary material The online version of this article (10.1186/s12864-019-5907-7) contains supplementary material, which is available to authorized users.

Recombination (R) rate and linkage disequilibrium (LD) analyses are the basis for plant breeding. These vary by breeding system, by generation of inbreeding or outcrossing and by region in the chromosome. Common bean (Phaseolus vulgaris L.) is a favored food legume with a small sequenced genome (514 Mb) and n = 11 chromosomes. The goal of this study was to describe R and LD in the common bean genome using a 768-marker array of single nucleotide polymorphisms (SNP) based on Trans-legume Orthologous Group (TOG) genes along with an advanced-generation Recombinant Inbred Line reference mapping population (BAT93 x Jalo EEP558) and an internationally available diversity panel. A whole genome genetic map was created that covered all eleven linkage groups (LG). The LGs were linked to the physical map by sequence data of the TOGs compared to each chromosome sequence of common bean. The genetic map length in total was smaller than for previous maps reflecting the precision of allele calling and mapping with SNP technology as well as the use of gene-based markers. A total of 91.4% of TOG markers had singleton hits with annotated Pv genes and all mapped outside of regions of resistance gene clusters. LD levels were found to be stronger within the Mesoamerican genepool and decay more rapidly within the Andean genepool. The recombination rate across the genome was 2.13 cM / Mb but R was found to be highly repressed around centromeres and frequent outside peri-centromeric regions. These results have important implications for association and genetic mapping or crop improvement in common bean.

Common bean (Phaseolus vulagaris L.) is mainly produced in Latin America, and Eastern and Southern Africa where seasonal rainfall is erratic and soil moisture deficit often limits its production. The objectives of this study were to identify superior advanced small red common bean lines with better grain yield under drought, and to identify plant traits that could serve as selection criteria for evaluating drought resistance. Thirty four advanced inbred lines were developed through triple cross and evaluated with two commercial checks (Nasir and Red Wolayita) under drought and irrigated field conditions in two seasons (2009 and 2010) and at two locations (Melkassa and Doni Kumbi Farm) in Ethiopia. Grain yield, pods/plant, seeds/plant and 100 seed weight were reduced by 47, 19, 15, and 9 %, respectively, under the drought conditions compared to the irrigated conditions. Seven genotypes (L62, L5, L20, L79, L12, L41, and L83) had better yield under drought compared to the standard check. Similarly, these genotypes had better predicted genotypic effects, suggesting that they would also be useful as drought resistant donors in common bean cultivar development program. Genotypic variance expressed as proportions to phenotypic variance was higher compared to genotype 9 environment interaction effects for all traits under drought conditions, suggesting that these genotypes can be used to further improve drought-resistance regardless of environments. Strong correlation between grain yield, canopy biomass and harvest index across environments suggested that canopy biomass and harvest index are repeatable and reliable plant traits to determine drought resistance in common bean.

Background Due to the recent domestication of peanut from a single tetraploidization event, relatively little genetic diversity underlies the extensive morphological and agronomic diversity in peanut cultivars today. To broaden the genetic variation in future breeding programs, it is necessary to characterize germplasm accessions for new sources of variation and to leverage the power of genome-wide association studies (GWAS) to discover markers associated with traits of interest. We report an analysis of linkage disequilibrium (LD), population structure, and genetic diversity, and examine the ability of GWA to infer marker-trait associations in the U.S. peanut mini core collection genotyped with a 58 K SNP array. Results LD persists over long distances in the collection, decaying to r 2 = half decay distance at 3.78 Mb. Structure within the collection is best explained when separated into four or five groups (K = 4 and K = 5). At K = 4 and 5, accessions loosely clustered according to market type and subspecies, though with numerous exceptions. Out of 107 accessions, 43 clustered in correspondence to the main market type subgroup whereas 34 did not. The remaining 30 accessions had either missing taxonomic classification or were classified as mixed. Phylogenetic network analysis also clustered accessions into approximately five groups based on their genotypes, with loose correspondence to subspecies and market type. Genome wide association analysis was performed on these lines for 12 seed composition and quality traits. Significant marker associations were identified for arachidic and behenic fatty acid compositions, which despite having low bioavailability in peanut, have been reported to raise cholesterol levels in humans. Other traits such as blanchability showed consistent associations in multiple tests, with plausible candidate genes. Conclusions Based on GWA, population structure as well as additional simulation results, we find that the primary limitations of this collection for GWAS are a small collection size, significant remaining structure/genetic similarity and long LD blocks that limit the resolution of association mapping. These results can be used to improve GWAS in peanut in future studies – for example, by increasing the size and reducing structure in the collections used for GWAS. Electronic supplementary material The online version of this article (10.1186/s12864-019-5824-9) contains supplementary material, which is available to authorized users.