Scholarlink

<p>Callogenesis is a complex process of development, the efficacy of which depends on the impact of exogenous and endogenous factors, and the most crucial aspect of the regulation is the plant growth regulators (PGRs). In <i>Clerodendrum phlomidis</i> L. f., callogenesis is a significant biotechnological method for the in vitro synthesis of secondary metabolites as well as the facilitation of indirect organogenesis. With the combination of machine learning (ML) and optimisation, the overall knowledge of the callogenesis and optimal protocol can be acquired. In the present study, the predictions of callogenesis responses (callus induction days and callus fresh weight) of <i>C. phlomidis</i> against various types and concentrations of PGRs were the indole-3-acetic acid (IAA), 1-naphthaleneacetic acid (NAA), 6-benzylaminopurine (BAP), and kinetin (Kn) at concentrations of 0.5–4.0&#xa0;mg/L with the multilayer perceptron (MLP), Support Vector machine (SVM), Random Forest (RF), Gradient Boosting Machine (GBM), and eXtreme Gradient Boosting (XGBoost). According to the findings, the XGBoost model was accurate in terms of predicting callogenesis responses of <i>C. phlomidis</i> (R<sup>2</sup> &gt; 0.85) on the training set and in the (R<sup>2</sup> &gt; 0.99) on the testing set. The genetic algorithm (GA) combined with the XGBoost model was used to optimise PGR concentrations to get maximum responses during in vitro callogenesis. The XGBoost-GA model’s predictive reliability was confirmed through experimental validation of the optimized parameters. Under these conditions, observed outcomes closely aligned with model prediction, yielding a callus induction period of 13.5 ± 0.3 days and a callus fresh weight of 683.26 ± 1.71&#xa0;mg. Overall, the results of the present investigation revealed that XGBoost-GA is a practical and robust predictive and optimisation method used in the in vitro callogenesis of <i>C. phlomidis</i>. The GC-MS detection had shown that in vitro callogenesis assists in the synthesis of specific phytochemicals.</p> Graphical Abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine learning-driven optimization of Clerodendrum phlomidis L. f. (sage glory bower) tissue culture for enhanced callogenesis

  • 期刊论文

  • 发表日期：

    2026年4月24日

查看全文
  • R. Praveen Kumar,
  • S. Ramesh Kumar

摘要

Callogenesis is a complex process of development, the efficacy of which depends on the impact of exogenous and endogenous factors, and the most crucial aspect of the regulation is the plant growth regulators (PGRs). In Clerodendrum phlomidis L. f., callogenesis is a significant biotechnological method for the in vitro synthesis of secondary metabolites as well as the facilitation of indirect organogenesis. With the combination of machine learning (ML) and optimisation, the overall knowledge of the callogenesis and optimal protocol can be acquired. In the present study, the predictions of callogenesis responses (callus induction days and callus fresh weight) of C. phlomidis against various types and concentrations of PGRs were the indole-3-acetic acid (IAA), 1-naphthaleneacetic acid (NAA), 6-benzylaminopurine (BAP), and kinetin (Kn) at concentrations of 0.5–4.0 mg/L with the multilayer perceptron (MLP), Support Vector machine (SVM), Random Forest (RF), Gradient Boosting Machine (GBM), and eXtreme Gradient Boosting (XGBoost). According to the findings, the XGBoost model was accurate in terms of predicting callogenesis responses of C. phlomidis (R2 > 0.85) on the training set and in the (R2 > 0.99) on the testing set. The genetic algorithm (GA) combined with the XGBoost model was used to optimise PGR concentrations to get maximum responses during in vitro callogenesis. The XGBoost-GA model’s predictive reliability was confirmed through experimental validation of the optimized parameters. Under these conditions, observed outcomes closely aligned with model prediction, yielding a callus induction period of 13.5 ± 0.3 days and a callus fresh weight of 683.26 ± 1.71 mg. Overall, the results of the present investigation revealed that XGBoost-GA is a practical and robust predictive and optimisation method used in the in vitro callogenesis of C. phlomidis. The GC-MS detection had shown that in vitro callogenesis assists in the synthesis of specific phytochemicals.

Graphical Abstract