Said Arturo Rodríguez-Romero

Pitahaya (Hylocereus spp.), commonly known as dragon fruit, has attracted increasing attention as a high-value crop because of its nutritional attributes, environmental adaptability, and growing market demand. However, existing reviews typically address individual topics—such as nutrition, cultivation, postharvest management, or sustainability—without integrating these aspects across the production cycle. This fragmentation hinders a comprehensive understanding of how propagation methods, agronomic and reproductive management, and cultivation conditions influence fruit quality and postharvest performance. This review synthesizes scientific and technological evidence on propagation, cultivation systems, agronomic management, nutritional composition, fruit quality, postharvest technologies, by-product valorization, sustainability, and innovation. Scientific publications and patents published between January 2019 and March 2026 were analyzed. An initial search identified approximately 170 scientific records and 2000 patent records; after screening, 105 scientific publications and 96 patents were included in the review. The evidence indicates advances in vegetative propagation, greenhouse and soilless cultivation, nutrient and environmental management, characterization of bioactive compounds, postharvest preservation, and by-product valorization. Patent activity also reflects technological developments in protected cultivation, structural support systems, and digital agriculture. Nevertheless, commercial-scale validation, standardized crop-specific management protocols, quantitative comparisons among production systems, and effective technology transfer remain limited.

Extracellular poly-γ-glutamic acid (γ-PGA) produced by Bacillus species demonstrates significant antibacterial properties, positioning it as a promising candidate for diverse biomedical and industrial applications. This study focused on molecular identification of Bacillus subtilis using Polymerase Chain Reaction (PCR) and evaluated the initial production of γ-PGA from a novel biological source of Bacillus subtilis. Shake flask fermentation was utilized for γ-PGA production, with three distinct growth media (Tryptic, MRS, and Mineral medium) assessed for their efficiency in polymer yield. Characterization of γ-PGA was conducted through FT-IR, HPLC, and GC-MS analyses. FT-IR spectroscopy confirmed the presence of characteristic functional groups such as carbonyl, amide, and hydroxyl groups. HPLC and GC-MS analyses provided insights into the polymer’s purity and molecular composition, highlighting components like methyl esters, hexanoic acid, and monomethyl esters. Furthermore, the study quantified γ-PGA production during a four-day shake flask fermentation period. These findings contribute significantly to bacterial characterization, optimization of fermentation processes, and the exploration of γ-PGA’s potential as an antibacterial agent. Future research directions include refining purification techniques to enhance γ-PGA’s antibacterial efficacy and expanding its applications across various fields.

This study presents a hybrid recurrent neural network (RNN) approach for greenhouse microclimate prediction, combining a mechanistic model with an Elman network. The research addresses the gap in systematic comparisons between hybrid RNN and feedforward neural network (FFNN) architectures for greenhouse climate forecasting. Different network structures with 1, 2, 3, 5, and 7 hidden layers were evaluated using mean absolute percentage error (MAPE), mean square error (MSE), and coefficient of determination (R²). Results demonstrate that hybrid RNNs significantly outperform FFNNs in predicting indoor temperature, with the 2-hidden-layer configuration achieving the best performance (R² = 0.897). For relative humidity prediction, both networks showed comparable results. The hybrid RNN with 3 hidden layers exhibited optimal performance during training, while simpler configurations proved more effective during testing. The integration of mechanistic knowledge with neural networks enhances prediction accuracy, providing a reliable tool for greenhouse climate control systems. These findings contribute to smart agriculture by offering an efficient computational approach for microclimate management.

The dehydration process modifies the physical and chemical characteristics of certain crops, thereby increasing their shelf life and consequently reducing the organic waste generated. This process is contingent upon maintaining optimal temperature and humidity levels to prevent deterioration of the product. As indirect dehydrators have a high energy demand, new designs are required that facilitate the uniform distribution of air with a high-volume capacity of 100 kg per day. In the present study, computational fluid dynamics (CFD) techniques were employed to assess the drying performance of two dehydrator models. The simulations were executed in Solidworks 2020 and Flow Simulation, and they examined temperature distribution and velocity within the interior of the dehydrators. In Model 1, an inlet volume flow of 0.08 m³ s⁻¹ and a heat source of 3.5 kW are considered, within a volume of 2.11 m³. In Model 2, an inlet volume flow of 0.03 m³ s⁻¹ and two heat source of 2.5 kW are considered, within a volume of 2.02 m³. Model 1 was unable to achieve uniform air distribution within the drying chamber. In contrast, Model 2 demonstrated uniform velocity and temperature across the majority of the drying chamber, making it a superior option.