Comparing quantum machine learning and classical machine learning for in vitro regeneration of cowpea (Vigna unguiculata)

Plant biotechnology is a key component of plant breeding which employs biotechnological tools like plant tissue culture for genetic engineering, genome editing, or other techniques. In this study, an efficient regeneration protocol was established followed by optimizing Benzylaminopurine (BAP) concentration for two different explants of cowpea was established. The data generated was analyzed by using traditional statistical tools like ANOVA and multiple factorial regression analysis. Results illustrated the better performance of both explants when pulse treated with 5.0 mg/L BAP followed by inoculation at 0.25 mg/L BAP or 0.0 BAP. Results were further analyzed and optimized by the Pareto chart analysis, surface plots, contour plots, and response optimizer. After that, data was computed with machine learning models [Support Vector Classifier (SVC), Random Forest (RF), and Multilayer Perceptron (MLP)] and compared with Quantum machine learning-based Quantum Support Vector Classifier (QSVC), and Variational Quantum Classifier (VQC) models. Results revealed better performance of the MLP model for most of the performance metrics. On the contrary, QSVC and VQC displayed better accuracy and recall respectively. Results depicted the promising performance of Quantum ML for better data analysis for decision-making in precision agriculture and precision biotechnology.Key message: The study integrates in vitro regeneration data into classical and quantum machine learning models, showing superior accuracy and recall in quantum models, highlighting its potential for decision-making in biotechnology. [ABSTRACT FROM AUTHOR]