Plant intelligence in silico: Integrating hybrid artificial intelligence and synthetic biology to decode adaptive plant strategies

Abstract

Plant intelligence is defined as the capacity of plants to integrate environmental signals through distributed biochemical, electrical, and epigenetic networks, resulting in adaptive physiological or developmental responses without requiring neural cognition. This definition distinguishes plant adaptive plasticity from animal cognition while maintaining a rigorous biological framework. This phenomenon has captivated biologists for decades, yet its underlying mechanisms remain elusive amid intricate biological interactions and vast, complex datasets. Recent breakthroughs in synthetic biology and artificial intelligence (AI) offer unprecedented opportunities to decode these adaptive strategies in silico. This review explores hybrid AI systems that integrates predictive AI, agent-based modelling, including large language models (LLMs) and retrieval-augmented generation (RAG), and bio-inspired algorithms, charting innovative pathways for simulating and investigating plant adaptive behaviours via synthetic biology. We highlight how these hybrid approaches bridge empirical science and theoretical insights, empowering the design of resilient synthetic biological systems that emulate plants' extraordinary adaptability to diverse challenges.