Optimizing Micropropagation and Rooting Protocols for Diverse Lavender Genotypes: A Synergistic Approach Integrating Machine Learning Techniques

Abstract

This study comprehensively explored the micropropagation and rooting capabilities of four distinct lavender genotypes, utilizing culture media with and without 2 g/L of activated charcoal. A systematic examination of varying concentrations of BAP for micropropagation and IBA for rooting identified an optimal concentration of 1 mg/L for both BAP and IBA, resulting in excellent outcomes. Following robust root development, the acclimatization of plants to external conditions achieved a 100% survival rate across all genotypes. In addition to the conventional techniques employed, integrating machine learning (ML) methodologies holds promise for further enhancing the efficiency of lavender propagation protocols. Using cutting-edge computational tools, including MLP, RBF, XGBoost, and GP algorithms, our findings were rigorously examined and forecast using three performance measures (RMSE, R2, and MAE). Notably, the comparative evaluation of different machine learning models revealed distinct R2 rates for plant characteristics, with MLP, RBF, XGBoost, and GP demonstrating varying degrees of effectiveness. Future studies may leverage ML models, such as XGBoost, MLP, RBF, and GP, to fine-tune specific variables, including culture media composition and growth regulator treatments. The adaptability and ability of ML techniques to analyze complex biological processes can provide valuable insights into optimizing lavender micropropagation on a broader scale. This collaborative approach, combining traditional in vitro techniques with machine learning, validates the success of current micropropagation and rooting protocols and paves the way for continuous improvement. By embracing ML in lavender propagation studies, researchers can contribute to advancing sustainable and efficient plant propagation techniques, thereby fostering the preservation and exploitation of genetic resources for conservation and agriculture.

Leveraging machine learning to unravel the impact of cadmium stress on goji berry micropropagation

“…This consistency across studies reinforces the versatility and reliability of RF in plant tissue culture applications, offering a robust framework for predictive modeling. The integration of machine learning (ML) algorithms in plant tissue culture, as explored by Şimşek et al [ 37 ] and further substantiated by our research findings, underscores the pivotal role computational methodologies are beginning to play in advancing plant science. Our study, utilizing algorithms such as MLP, SVM, RF, GP, and XGBoost, shares a common narrative with Şimşek et al [ 37 ] in harnessing these tools to refine predictions on plant growth outcomes under varying conditions, such as Cd stress.…”

“…The integration of machine learning (ML) algorithms in plant tissue culture, as explored by Şimşek et al [ 37 ] and further substantiated by our research findings, underscores the pivotal role computational methodologies are beginning to play in advancing plant science. Our study, utilizing algorithms such as MLP, SVM, RF, GP, and XGBoost, shares a common narrative with Şimşek et al [ 37 ] in harnessing these tools to refine predictions on plant growth outcomes under varying conditions, such as Cd stress. In both instances, the differential effectiveness of these ML models highlights the importance of strategic model selection tailored to the specific objectives of the research.…”

“…Numerous ML models effectively forecast and refine plant tissue culture procedures across various studies [ 32 – 37 ]. ML has demonstrated significant potential in enhancing various aspects of in vitro culture systems, providing improvements over traditional statistical methods.…”

Saffron In Vitro Propagation: An Innovative Method by Temporary Immersion System (TIS), Integrated with Machine Learning Analysis

“…The input and output variables from the training set were used to train the MLP using a supervised training technique. Until the desired value in Equation ( 4) was attained, the training procedure was repeated [26].…”

“…Many researchers frequently find it challenging to use conventional statistical methods to assess big and complex datasets in the context of in vitro micropropagation, a composite biological process impacted by genotypes, culture medium, and environment [26]. Recently, new technology based on artificial intelligence, such as machine learning, is developing quickly in several scientific and industrial domains [27] but its integration within the plant and agricultural sciences remains relatively emergent.…”