Search

The current investigation addresses the pressing need to integrate orphan or underutilized crops into mainstream agriculture, focusing on Perilla (Perilla frutescens L.) due to its superior nutritional profile. A major challenge is the lack of fast, cost-effective, and labor-efficient screening methods for germplasm. Near-Infrared Reflectance Spectroscopy (NIRS) addresses this by providing precise and rapid determination of crucial biochemical parameters. This study developed Modified Partial Least Squares (mPLS) regression-based NIRS prediction models using WinISI and 1D Convolutional Neural Networks (CNN) to enable high-throughput screening for moisture, ash, proteins, total soluble sugars (TSS), and phenols in Perilla germplasm. Calibration with WinISI involved mathematical treatments, optimizing for each trait: "2,6,6,1" for moisture, "3,4,4,1" for ash and TSS, "3,4,6,1" for protein, and "2,4,6,1" for phenols. The 1D CNN model, with lower mean absolute error (MAE), was further validated. External validation metrics, including RSQexternal, SEP(C), slope, bias, and RPD, assessed prediction accuracy. Comparative evaluation showed WinISI performed better for moisture prediction, while the 1D CNN model excelled in predicting ash, protein, TSS, and total phenol, highlighting the importance of model selection for specific traits. This rapid screening tool aids in identifying nutritionally dense Perilla genotypes, guiding targeted breeding efforts, and represents the first comparative mPLS and DL-based modeling using NIRS data for Perilla. [ABSTRACT FROM AUTHOR]

Potato is a globally significant crop, crucial for food security and nutrition. Assessing vital nutritional traits is pivotal for enhancing nutritional value. However, traditional wet lab methods for the screening of large germplasms are time- and resource-intensive. To address this challenge, we used near-infrared reflectance spectroscopy (NIRS) for rapid trait estimation in diverse potato germplasms. It employs molecular absorption principles that use near-infrared sections of the electromagnetic spectrum for the precise and rapid determination of biochemical parameters and is non-destructive, enabling trait monitoring without sample compromise. We focused on modified partial least squares (MPLS)-based NIRS prediction models to assess eight key nutritional traits. Various mathematical treatments were executed by permutation and combinations for model calibration. The external validation prediction accuracy was based on the coefficient of determination (RSQexternal), the ratio of performance to deviation (RPD), and the low standard error of performance (SEP). Higher RSQexternal values of 0.937, 0.892, and 0.759 were obtained for protein, dry matter, and total phenols, respectively. Higher RPD values were found for protein (3.982), followed by dry matter (3.041) and total phenolics (2.000), which indicates the excellent predictability of the models. A paired t-test confirmed that the differences between laboratory and predicted values are non-significant. This study presents the first multi-trait NIRS prediction model for Indian potato germplasm. The developed NIRS model effectively predicted the remaining genotypes in this study, demonstrating its broad applicability. This work highlights the rapid screening potential of NIRS for potato germplasm, a valuable tool for identifying trait variations and refining breeding strategies, to ensure sustainable potato production in the face of climate change. [ABSTRACT FROM AUTHOR]

Perilla seed meal (PSM), a byproduct of oil extraction from Perilla frutescens L. seeds, is rich in protein (24.26–42.85%) and holds potential as an economical and sustainable animal feed. Traditional methods for assessing protein content are labor-intensive and costly. This study explores Near-Infrared Reflectance Spectroscopy (NIRS) for the rapid, precise, and non-destructive determination of PSM protein content in 126 samples. We developed and evaluated Modified Partial Least Squares (MPLS) regression and deep learning (DL) models, including 1D-CNN (Convolutional Neural Network), LSTM Long Short-Term Memory), and hybrid architectures incorporating skip connections, inception modules, and spectral derivatives. Model performance was validated externally using parameters such as RSQ external (R-squared), bias, SEP(C) (Standard Error of Prediction), RPD (Residual Prediction Deviation), slope, SD (Standard Deviation), p-value (≥0.05), and the correlation between reference and predicted values. The 1D CNN-LSTM-Inception derivative 1 model achieved the best performance (RPD: 8.0, RSQ external : 0.98), followed by the MPLS-based model (RPD: 4.88, RSQ external : 0.96) and the 1D CNN derivative 1 model (RPD: 3.07, RSQ external : 0.96). These models provide a reliable and advanced technology for the non-destructive screening of PSM protein content, thus aiding in the rapid identification and selection of superior perilla chemotypes from varied backgrounds. [ABSTRACT FROM AUTHOR]

Perilla seeds contain a rich array of essential minerals, thus having the potential to address multiple micronutrient deficiencies at a time. However, traditional methods of mineral estimation are complex, time-consuming, expensive, and require technical expertise. This study includes the development of Near-Infrared Reflectance Spectroscopy (NIRS)-based prediction models for predicting five important minerals (Calcium, Copper, Magnesium, Manganese, and Phosphorus) using machine learning and deep learning techniques. Four models, including 1D Convolutional Neural Networks (1D CNNs), Artificial Neural Networks (ANNs), Random Forests (RFs), and Support Vector Regression (SVR), were developed and evaluated. The developed 1D CNN model outperformed other considered models in predicting calcium, magnesium, and phosphorus content with RPD (Residual Prediction Deviation) values of 1.75, 1.83, and 2.96, respectively. Whereas, SVR performed best in predicting copper and manganese with an RPD of 1.82 and 2.2, respectively. The 1D CNN model demonstrated R2 (Coefficient of determination) values above 0.65 for all minerals, with a maximum of 0.88 for phosphorus. In addition, the developed models performed superior as compared to the Partial Least Square Regression method (R2= 0.32). The developed models provide efficient tools for rapidly screening perilla germplasm available in global repositories, thus aiding in the selection of mineral-rich genotypes to mitigate micronutrient deficiencies. • Predicted minerals content in Perilla seeds. • Near-infrared reflectance spectroscopy-based models were developed. • Machine learning and deep learning-based techniques were applied. • Deep learning-based model achieved R2 value above 0.65 for all minerals. [ABSTRACT FROM AUTHOR]

[Display omitted] • Maiden use of Transformers in nutritional profiling of agricultural products. • Used Transformers for protein prediction using NIRS of Lablab purpureus L. • Transformer model achieved highest predictive accuracy (RSQ = 0.977, RPD = 13.92). • Transformer models outperformed CNN by 22.86% and MPLS by 32.21% in RPD. Lablab bean (Lablab purpureus L.), known for its higher protein content provides a promising alternative to reduce reliance on animal-based proteins and support sustainable agriculture. Nowadays, traditional methods for nutritional profiling have been outperformed by Convolutional Neural Networks (CNNs) integrated with Near-Infrared Reflectance Spectroscopy (NIRS). However, an advanced technique called Transformers, despite their potential, remains underexplored for prediction of key quality traits of crops. Thus, in the present study, we aimed to predict lablab bean protein content using a Transformer-based model (termed ProTformer) coupled with NIRS. The Dumas combustion method was used to determine protein content from 112 lablab bean genotypes. The performance of this model was compared with 1D CNN and Modified Partial Least Squares (MPLS)-based models. These models were evaluated using the Coefficient of Determination (RSQ), Residual Prediction Deviation (RPD), bias, and Corrected Standard Error of Prediction (SEP(C)). Our findings show that the Transformer-based model achieved the highest predictive accuracy (RSQ = 0.977, RPD = 13.92), surpassing the 1D CNN (RSQ = 0.956, RPD = 11.33) and MPLS (RSQ = 0.943, RPD = 10.53)- based models by 22.86% and 32.21% in RPD, respectively. The multi-head self-attention mechanism present in the Transformer-based model enables it to concurrently focus on different spectral regions, resulting in superior predictive performance. This study shows the potential of using Transformer-based models coupled with NIRS for nutritional profiling which could be effectively adapted for other crops for rapid screening of large germplasm available at global repositories. [ABSTRACT FROM AUTHOR]

Rice bean, a potential protein-rich underutilized crop, can serve as a valuable source for food and feed, contributing to sustainable food and nutrition security. However, traditional methods for measuring protein content are laborious, tedious, and expensive. This study evaluates the performance of Modified Partial Least Squares (MPLS) and 1D Convolutional Neural Network (1D CNN) models in predicting protein content in rice bean using Near-Infrared Reflectance Spectroscopy (NIRS) data. Both models were validated with whole spectra, key wavelengths, and spectral segments derived from Genetic Algorithm (GA), Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO), and Competitive Adaptive Reweighted Sampling (CARS). The 1D CNN model demonstrated superior predictive performance, with an 11 % increase in R² and an 18 % increase in RPD using the whole spectra compared to the MPLS model. Specifically, the 1D CNN model with CARS-selected wavelengths achieved an R² of 0.82 and an RPD of 2.15, while the MPLS model achieved an R² of 0.84 and an RPD of 2.25. The segmented spectra approach further enhanced the 1D CNN model's performance, achieving the highest R² of 0.84 and an RPD of 2.28 with CARS-selected segments. These results highlight that the 1D CNN model with CARS-selected wavelengths provides the best predictive performance for protein content in rice bean. The optimized method can lead to rapid early-phase breeding practices and improved quality assessments in leguminous crops, emphasizing the importance of integrating advanced modeling techniques with NIRS for agricultural applications. • Evaluated MPLS and 1D CNN models to predict rice bean protein content using NIRS. • Used metaheuristic algorithms GA, ACO, PSO, and CARS for wavelength selection. • Compared whole spectra, key wavelengths, and segmented spectra approaches. • 1D CNN outperformed MPLS with 18 % higher RPD using whole spectra. • 1D CNN with CARS-selected segments achieved highest R² of 0.84 and RPD of 2.28. [ABSTRACT FROM AUTHOR]

Mung bean is considered one of the significant protein sources with high digestibility; a wide range is observed for protein content in germplasm collections. Near-infrared reflectance spectroscopy (NIRS) is a rapid and non-destructive analytical technique widely used for predicting protein and other nutrients. One hundred and fifty germplasm with a wide range of protein contents (19.7–29.1 %) were selected to build the model. A set of 100 accessions was used for training and validated on the remaining 50 accessions. A systematic comparison of modified partial least squares (mPLS) and partial least squares (PLS) regression techniques with different pre-processing methods (SNV-DT and MSC) and mathematical treatments were carried out. Validation of the calibration equation returned the highest R 2 value of 0.940 with mathematical treatment 2882 using mPLS regression, and 0.934 with mathematical treatment 2442 for PLS regression, having ratio prediction deviations (RPD) of 3.84 and 3.59, respectively. Based on the external validation results, the mPLS regression method was found to be superior to the PLS method. • Efficient prediction model for protein content range of 19.7–29.1 % in mung bean. • NIR spectra pre-processing has maximum influence on prediction model performance. • Modified partial least squares (mPLS), partial least squares (PLS) with SNV detrend were compared. • mPLS was found to be superior to PLS with fewer outliers and better R2 values. • Best prediction model for protein observed with mathematical treatment of 2882. [ABSTRACT FROM AUTHOR]