Utilization of Synthetic Near-Infrared Spectra via Generative Adversarial Network to Improve Wood Stiffness Prediction.

Near-infrared (NIR) spectroscopy is widely used as a nondestructive evaluation (NDE) tool for predicting wood properties. When deploying NIR models, one faces challenges in ensuring representative training data, which large datasets can mitigate but often at a significant cost. Machine learning and deep learning NIR models are at an even greater disadvantage because they typically require higher sample sizes for training. In this study, NIR spectra were collected to predict the modulus of elasticity (MOE) of southern pine lumber (training set = 573 samples, testing set = 145 samples). To account for the limited size of the training data, this study employed a generative adversarial network (GAN) to generate synthetic NIR spectra. The training dataset was fed into a GAN to generate 313, 573, and 1000 synthetic spectra. The original and enhanced datasets were used to train artificial neural networks (ANNs), convolutional neural networks (CNNs), and light gradient boosting machines (LGBMs) for MOE prediction. Overall, results showed that data augmentation using GAN improved the coefficient of determination (R2) by up to 7.02% and reduced the error of predictions by up to 4.29%. ANNs and CNNs benefited more from synthetic spectra than LGBMs, which only yielded slight improvement. All models showed optimal performance when 313 synthetic spectra were added to the original training data; further additions did not improve model performance because the quality of the datapoints generated by GAN beyond a certain threshold is poor, and one of the main reasons for this can be the size of the initial training data fed into the GAN. LGBMs showed superior performances than ANNs and CNNs on both the original and enhanced training datasets, which highlights the significance of selecting an appropriate machine learning or deep learning model for NIR spectral-data analysis. The results highlighted the positive impact of GAN on the predictive performance of models utilizing NIR spectroscopy as an NDE technique and monitoring tool for wood mechanical-property evaluation. Further studies should investigate the impact of the initial size of training data, the optimal number of generated synthetic spectra, and machine learning or deep learning models that could benefit more from data augmentation using GANs.

Compensatory Guaiacyl Lignin Biosynthesis at the Expense of Syringyl Lignin in 4CL1-Knockout Poplar.

The lignin biosynthetic pathway is highly conserved in angiosperms, yet pathway manipulations give rise to a variety of taxon-specific outcomes. Knockout of lignin-associated 4-coumarate:CoA ligases (4CLs) in herbaceous species mainly reduces guaiacyl (G) lignin and enhances cell wall saccharification. Here we show that CRISPR-knockout of 4CL1 in poplar (Populus tremula × alba) preferentially reduced syringyl (S) lignin, with negligible effects on biomass recalcitrance. Concordant with reduced S-lignin was downregulation of ferulate 5-hydroxylases (F5Hs). Lignification was largely sustained by 4CL5, a low-affinity paralog of 4CL1 typically with only minor xylem expression or activity. Levels of caffeate, the preferred substrate of 4CL5, increased in line with significant upregulation of caffeoyl shikimate esterase1 Upregulation of caffeoyl-CoA O-methyltransferase1 and downregulation of F5Hs are consistent with preferential funneling of 4CL5 products toward G-lignin biosynthesis at the expense of S-lignin. Thus, transcriptional and metabolic adaptations to 4CL1-knockout appear to have enabled 4CL5 catalysis at a level sufficient to sustain lignification. Finally, genes involved in sulfur assimilation, the glutathione-ascorbate cycle, and various antioxidant systems were upregulated in the mutants, suggesting cascading responses to perturbed thioesterification in lignin biosynthesis.

A quantitative method for analyzing glycome profiles of plant cell walls.

Glycome profiling allows for the characterization of plant cell wall ultrastructure via sequential extractions and subsequent detection of specific epitopes with a suite of glycan-specific monoclonal antibodies (mAbs). The data are often viewed as the amount of materials recovered and coinciding colored heatmaps of mAb binding are generated. Interpretation of these data can be considered qualitative in nature as it depends on detecting subtle visual differences in antibody binding strength. Here, we report a mixed model-based quantitative approach for glycome profile analyses, which accounts for the amount of materials recovered and displays the normalized values in revised heatmaps and statistical heatmaps depicting significant differences. The utility of this methodology was demonstrated on a previously published dataset investigating the effects of moisture stress on the roots and needles of Pinus taeda. An annotated R script for the quantitative methodology is included to allow future studies to utilize the same approach.