Biology subjects

Ganapathysubramanian, B.

Publications and source records attributed to Ganapathysubramanian, B..

10 recordsLinked to original sources

bioRxiv2025

AI-assisted Image-Based Phenotyping Reveals Genetic Architecture of Pod Traits in Mungbean (Vigna radiata L.)

Mungbean (Vigna radiata (L.) R. Wilczek) is a vital source of digestible proteins and is well-suited for the plant-based protein industry. In this study, we analyzed pod morphological traits in the Iowa Mungbean Diversity (IMD) panel with 372 genotypes (2022-23) with AI-assisted image phenotyping using 2,418 pod images. Pod morphological traits were extracted using deep learning image analysis, achieving excellent agreement with manual measurements (r>0.96 for pod length and seed per pod). Four complementary GWAS models identified 45 significant SNPs associated with pod curvature, length, width, and seed per pod traits. Notably, a significant SNP (5_35265704) on chromosome 1 was linked to pod dimensional traits, length, width, and curvature. A candidate gene, Vradi01g00001116, was located within the linkage disequilibrium (LD) region of this SNP, is part of the GH3 gene family, and has an Arabidopsis ortholog (AT4G27260) known for influencing organ elongation, pod, and seed development. Another SNP, 5_210437 on chromosome 2, has been found to be significantly associated with both pod length and seed per pod. A candidate gene, Vradi02g00003971, located in the LD region of this SNP, belongs to the potassium transporter family and shares homology with the HAK5 gene family (AT4G13420) in Arabidopsis, which influences pod and seed growth. Image-based measurements achieved genomic prediction accuracies ranging from 0.61 to 0.85 across various traits, exhibiting an improvement of 12-22% over manual methods. These results demonstrate the potential of AI-assisted phenomics integrated with genomic tools to accelerate selection for improved pod architecture in mungbean breeding programs across the Midwestern United States and globally.

genomics↗

bioRxiv2025

Time Series GWAS for Iron Deficiency Chlorosis Tolerance in Soybean using Aerial Imagery

AbstractThe use of drones has become a commonly used tool by plant scientists to aid in plant phenotyping endeavors. Iron deficiency chlorosis (IDC) is a commonly observed abiotic stress in soybean fields with high soil pH levels. IDC severity is visually classified, and recent work has shown that digital imaging techniques using both ground and UAS-acquired imagery can be utilized for automated severity ratings. In our study, we compared the classification accuracy of two flight altitudes to determine the optimal flight parameters for IDC phenotyping. In addition to this, we investigated the ability to use image-predicted scores for genome wide association study (GWAS), as well as the effect of IDC on traits such as canopy area and canopy growth and development. We also report a tool for semi-automated plot extraction from orthomosaic images that can be easily integrated with UAS. We noted that 43 days after planting was an ideal time for IDC severity ratings as the highest number of significant SNPs were reported at this timepoint.

genetics↗

bioRxiv2024

Genomic and Phenomic Prediction for Soybean Seed Yield, Protein, and Oil

Developments in genomics and phenomics have provided valuable tools for use in cultivar development. Genomic prediction (GP) has been used in commercial soybean [Glycine max L. (Merr.)] breeding programs to predict grain yield and seed composition traits. Phenomic prediction (PP) is a rapidly developing field that holds the potential to be used for the selection of genotypes early in the growing season. The objectives of this study were to compare the use and performance of GP and PP for predicting soybean seed yield, protein content, and oil content. We additionally conducted Genome Wide Association Studies (GWAS) to identify significant SNPs associated with the traits of interest. These SNPs were also used to train the GP models. The GWAS panel of 292 diverse accessions was grown in six environments in replicated trials. Spectral data were collected at three timepoints during the growing season. A GBLUP model was trained on 268 accessions, while three separate machine learning (ML) models were trained on vegetation indices (VIs) and canopy traits. We observed that for PP, Random Forest (RF) algorithm had the highest rank correlation between the predicted and the actual phenotype rank. PP had a higher correlation coefficient than GP for seed yield, while GP had higher correlation coefficients for seed protein and oil contents. VIs with high feature importance were used as covariates in a new GBLUP model, and a new RF model was trained with the inclusion of selected SNPs from the GWAS results. These models did not outperform the original GP and PP models. These results show the capability of using ML for in-season predictions for specific traits in soybean breeding and provide insights on PP and GP inclusions in breeding programs. PLAIN LANGUAGE SUMMARYObtaining DNA information can often be costly for a public breeding program to obtain, which is a barrier of entry for using genomic prediction in these programs to make selections. Phenomic prediction provides an alternative opportunity that does not require the recurring costs of DNA sequencing. This research aimed to compare genomic and phenomic prediction in a diverse panel of soybeans. We found that phenomic prediction had the highest accuracy for seed yield, while genomic prediction was the best model for seed protein and oil. Combining genomic and phenomic data did not improve the predictive ability of the models. Core ideas O_LIIn-season phenomic prediction is able to outperform genomic prediction for seed yield. C_LIO_LIGenomic prediction outperformed in-season phenomic prediction models for seed protein and oil. C_LIO_LIOverlapping spectral indices were identified as the most predictive for seed yield, protein, and oil. C_LIO_LIFusion of genomic and phenomic prediction did not increase the predictive ability of the combined models. C_LI

plant biology↗

bioRxiv2024

Leveraging Soil Mapping and Machine Learning to Improve Spatial Adjustments in Plant Breeding Trials

Spatial adjustments are used to improve the estimate of plot seed yield across crops and geographies. Moving mean and P-Spline are examples of spatial adjustment methods used in plant breeding trials to deal with field heterogeneity. Within trial spatial variability primarily comes from soil feature gradients, such as nutrients, but study of the importance of various soil factors including nutrients is lacking. We analyzed plant breeding progeny row and preliminary yield trial data of a public soybean breeding program across three years consisting of 43,545 plots. We compared several spatial adjustment methods: unadjusted (as a control), moving means adjustment, P-spline adjustment, and a machine learning based method called XGBoost. XGBoost modeled soil features at (a) local field scale for each generation and per year, and (b) all inclusive field scale spanning all generations and years. We report the usefulness of spatial adjustments at both progeny row and preliminary yield trial stages of field testing, and additionally provide ways to utilize interpretability insights of soil features in spatial adjustments. These results empower breeders to further refine selection criteria to make more accurate selections, and furthermore include soil variables to select for macro- and micro-nutrients stress tolerance.

plant biology↗

bioRxiv2023

Data driven discovery and quantification of hyperspectral leaf reflectance phenotypes across a maize diversity panel

Hyperspectral reflectance data can be collected from large plant populations in a high-throughput manner in both controlled and field environments. The efficacy of using hyperspectral leaf reflectance as a proxy for traits that typically require significant labor and time to collect has been evaluated in a number of studies. Commonly, estimating plant traits using hyperspectral reflectance involves collecting substantial amounts of ground truth data from plant populations, which may not be feasible for many researchers. In this study, we explore the potential of data-driven approaches to analyze hyperspectral reflectance data with little to no ground truth phenotypic measurements. Evaluations were performed using data on the reflectance of 2,151 individual wavelengths of light from the leaves of maize plants harvested from 1,658 field plots of a replicated trial including representatives of 752 maize genotypes from the Wisconsin Diversity Panel. We reduced the dimensionality of this dataset using an autoencoder neural network and principal component analyses, producing 10 latent variables and principal components, respectively. A subset of these principal components and latent variables demonstrated significant repeatability, indicating that a substantial proportion of the total variance in these variables was explained by genetic factors. Moreover, correlations were observed between variables derived from the autoencoder network and principal components with molecular traits. Notably, the most relevant latent variable (LV8) showed a much stronger correlation with chlorophyll content (R2 = 0.59) compared to the most correlated principal component (PC2; R2 = 0.31). Furthermore, one latent variable exhibited modestly better performance than a partial least squares regression model in estimating leaf chlorophyll content (PLSR; R2 = 0.58, LV8; R2 = 0.59). A number of genetic markers in the maize genome were significantly correlated with variation in different latent variables in genome wide association studies. In a number of cases, significant signals in genome wide association studies were adjacent to genes with plausible links to traits expected to influence leaf hyperspectral reflectance patterns.

plant biology↗

bioRxiv2022

Self-Supervised Maize Kernel Classification and Segmentation for Embryo Identification

Computer vision and deep learning (DL) techniques have succeeded in a wide range of diverse fields. Recently, these techniques have been successfully deployed in plant science applications to address food security, productivity, and environmental sustainability problems for a growing global population. However, training these DL models often necessitates the large-scale manual annotation of data which frequently becomes a tedious and time-and-resource-intensive process. Recent advances in self-supervised learning (SSL) methods have proven instrumental in overcoming these obstacles, using purely unlabeled datasets to pre-train DL models. Here, we implement the popular self-supervised contrastive learning methods of NNCLR (Nearest neighbor Contrastive Learning of visual Representations) and SimCLR (Simple framework for Contrastive Learning of visual Representations) for the classification of spatial orientation and segmentation of embryos of maize kernels. Maize kernels are imaged using a commercial high-throughput imaging system. This image data is often used in multiple downstream applications across both production and breeding applications, for instance, sorting for oil content based on segmenting and quantifying the scutellums size and for classifying haploid and diploid kernels. We show that in both classification and segmentation problems, SSL techniques outperform their purely supervised transfer learning-based counterparts and are significantly more annotation efficient. Additionally, we show that a single SSL pre-trained model can be efficiently finetuned for both classification and segmentation, indicating good transferability across multiple downstream applications. Segmentation models with SSL-pretrained backbones produce DICE similarity coefficients of 0.81, higher than the 0.78 and 0.73 of those with ImageNet-pretrained and randomly initialized backbones, respectively. We observe that finetuning classification and segmentation models on as little as 1% annotation produces competitive results. These results show SSL provides a meaningful step forward in data efficiency with agricultural deep learning and computer vision.

plant biology↗

bioRxiv2020

Using Machine Learning To Develop A Fully Automated Soybean Nodule Acquisition Pipeline (SNAP)

Nodules form on plant roots through the symbiotic relationship between soybean (Glycine max L. Merr.) roots and bacteria (Bradyrhizobium japonicum), and are an important structure where atmospheric nitrogen (N2) is fixed into bio-available ammonia (NH3) for plant growth and developmental. Nodule quantification on soybean roots is a laborious and tedious task; therefore, assessment is done on a less informative qualitative scale. We report the Soybean Nodule Acquisition Pipeline (SNAP) for nodule quantification that combines RetinaNet and UNet deep learning architectures for object (i.e., nodule) detection and segmentation. SNAP was built using data from 691 unique roots from diverse soybean genotypes, vegetative growth stages, and field locations; and has a prediction accuracy of 99%. SNAP reduces the human labor and inconsistencies of counting nodules, while acquiring quantifiable traits related to nodule growth, location and distribution on roots. The ability of SNAP to phenotype nodules on soybean roots at a higher throughput enables researchers to assess the genetic and environmental factors, and their interactions on nodulation from an early development stage. The application of SNAP in research and breeding pipelines may lead to more nitrogen use efficient soybean and other legume species cultivars, as well as enhanced insight into the plant-Bradyrhizobium relationship.

plant biology↗

bioRxiv2018

Measurements of maize root plasticity under water stress in hydroponic chamber

1.Under water stress, plants adjust root traits including depth of root system, root diameter, density of root per volume of soil, hydraulic conductance of root. In this experimental study, we present a method to quantify how hydraulic traits of maize roots adapt to drought. The experiments involve a microfluidic flow sensor and a custom-built pressure chamber, made of transparent plastic for visualization purposes. We measured how maize genotypes (PHB47 and PHZ51) grown for a week in deionized (DI) water and one day in hydroponic nutrients solution (called the irrigated condition) respond to one week of water stress. Conditions of water stress (called drought conditions) were created by mixing Polyethylene Glycol with the nutrients solution. Results show that under drought, the roots of both genotypes respond by approximately halving their global hydraulic conductance. This adjustment seems to be achieved mainly by reductions of the total surface area of the roots. Interestingly, the measured hydraulic conductivity of the roots grown under drought was significantly larger. In all, this study sheds light on how plants adapt to water stress in a hydroponic system, by decreasing root area and increasing root permeability.

physiology↗

bioRxiv2018

Integrating genotype and weather variables for soybean yield prediction using deep learning

Realized performance of complex traits is dependent on both genetic and environmental factors, which can be difficult to dissect due to the requirement for multiple replications of many genotypes in diverse environmental conditions. To mediate these problems, we present a machine learning framework in soybean (Glycine max (L.) Merr.) to analyze historical performance records from Uniform Soybean Tests (UST) in North America, with an aim to dissect and predict genotype response in multiple envrionments leveraging pedigree and genomic relatedness measures along with weekly weather parameters. The ML framework of Long Short Term Memory - Recurrent Neural Networks works by isolating key weather events and genetic interactions which affect yield, seed oil, seed protein and maturity enabling prediction of genotypic responses in unseen environments. This approach presents an exciting avenue for genotype x environment studies and enables prediction based systems. Our approaches can be applied in plant breeding programs with multi-environment and multi-genotype data, to identify superior genotypes through selection for commercial release as well as for determining ideal locations for efficient performance testing.

bioinformatics↗

bioRxiv2018

Crowdsourcing Image Analysis for Plant Phenomics to Generate Ground Truth Data for Machine Learning

The accuracy of machine learning tasks critically depends on high quality ground truth data. Therefore, in many cases, producing good ground truth data typically involves trained professionals; however, this can be costly in time, effort, and money. Here we explore the use of crowdsourcing to generate a large number of training data of good quality. We explore an image analysis task involving the segmentation of corn tassels from images taken in a field setting. We investigate the accuracy, speed and other quality metrics when this task is performed by students for academic credit, Amazon MTurk workers, and Master Amazon MTurk workers. We conclude that the Amazon MTurk and Master Mturk workers perform significantly better than the for-credit students, but with no significant difference between the two MTurk worker types. Furthermore, the quality of the segmentation produced by Amazon MTurk workers rivals that of an expert worker. We provide best practices to assess the quality of ground truth data, and to compare data quality produced by different sources. We conclude that properly managed crowdsourcing can be used to establish large volumes of viable ground truth data at a low cost and high quality, especially in the context of high throughput plant phenotyping. We also provide several metrics for assessing the quality of the generated datasets.\n\nAuthor SummaryFood security is a growing global concern. Farmers, plant breeders, and geneticists are hastening to address the challenges presented to agriculture by climate change, dwindling arable land, and population growth. Scientists in the field of plant phenomics are using satellite and drone images to understand how crops respond to a changing environment and to combine genetics and environmental measures to maximize crop growth efficiency. However, the terabytes of image data require new computational methods to extract useful information. Machine learning algorithms are effective in recognizing select parts of images, but they require high quality data curated by people to train them, a process that can be laborious and costly. We examined how well crowdsourcing works in providing training data for plant phenomics, specifically, segmenting a corn tassel - the male flower of the corn plant - from the often-cluttered images of a cornfield. We provided images to students, and to Amazon MTurkers, the latter being an on-demand workforce brokered by Amazon.com and paid on a task-by-task basis. We report on best practices in crowdsourcing image labeling for phenomics, and compare the different groups on measures such as fatigue and accuracy over time. We find that crowdsourcing is a good way of generating quality labeled data, rivaling that of experts.

plant biology↗