Department of Medicine, University of The East, Manila, Philippines.
*Corresponding author: Fenella Chadwick
Department of Medicine, University of The East, Manila, Philippines.
Email ID: fenellachadwick@gmail.com
Received: Feb 21, 2025
Accepted: Mar 19, 2025
Published Online: Mar 26, 2025
Journal: Annals of Traditional Chinese Medicine
Copyright: Chadwick F. © All rights are reserved
Citation: Chadwick F. AI-driven drug discovery from Traditional Chinese Medicine: Accelerating the development of novel therapeutics. Ann Tradit Chin Med. 2025; 1(1): 1011.
Traditional Chinese Medicine (TCM) has a rich history of utilizing natural products for therapeutic purposes, offering a vast library of potential drug candidates. However, the complex nature of TCM formulas and the lack of systematic investigation have hindered the efficient discovery and development of novel therapeutics. Artificial Intelligence (AI) is emerging as a powerful tool to accelerate this process by leveraging the vast amount of TCM data, including herbal properties, formulas, and clinical applications. This paper explores the application of AI in various stages of TCM drug discovery, including target identification, compound screening, mechanism of action prediction, and drug repurposing. We discuss how machine learning, deep learning, and network pharmacology approaches can be used to analyze TCM data, identify active ingredients, predict drug targets, and elucidate the complex interactions within TCM formulas. Furthermore, we address the challenges and opportunities associated with AI-driven TCM drug discovery, such as data standardization, model validation, and the integration of TCM knowledge with modern drug discovery pipelines. This paper argues that AI has the potential to significantly accelerate the development of novel therapeutics derived from TCM, offering new avenues for treating a wide range of diseases.
Keywords: Artificial intelligence; Traditional chinese medicine; Drug discovery; Drug development; Machine learning; Deep learning; Network pharmacology; Herbal medicine; Natural products; Target identification; Compound screening; Mechanism of action; Drug repurposing; Data standardization; AI in drug discovery.
Traditional Chinese Medicine (TCM), with its origins stretching back millennia, represents a vast repository of knowledge regarding the use of natural products for therapeutic purposes. For centuries, TCM practitioners have relied on complex herbal formulas to treat a wide range of ailments, demonstrating the efficacy of these remedies [1-6] through empirical observation and clinical experience. This rich history offers a unique and valuable resource for modern drug discovery, providing a library of potentially bioactive compounds with established therapeutic applications. However, the complexity of TCM formulas, often containing multiple herbs with synergistic effects, and the lack of systematic investigation using modern scientific methods have hindered the efficient identification of active ingredients and the elucidation of their mechanisms of action. Furthermore, the vast amount of TCM literature, often written in classical Chinese, poses a significant challenge for researchers seeking to extract and analyze relevant information. Consequently, the translation of TCM knowledge into modern drug discovery pipelines has been a slow and arduous process.
The advent of Artificial Intelligence (AI) offers a transformative opportunity to accelerate the discovery and development of novel therapeutics derived from TCM. AI [7-10], with its ability to process vast amounts of data, identify complex patterns, and make predictions, is revolutionizing various fields, including drug discovery. In the context of TCM, AI can be leveraged to analyze the wealth of information contained in TCM texts, herbal databases, and experimental data to identify potential drug targets, screen compounds for activity, predict mechanisms of action, and even repurpose existing drugs for new TCM-related indications. By integrating AI with modern drug discovery techniques, researchers can significantly reduce the time and cost associated with developing new therapeutics from TCM.
AI-driven drug discovery from TCM involves several key steps. First, AI algorithms can be used to analyze TCM texts and databases to identify herbs and formulas with specific therapeutic properties. Natural Language Processing (NLP) techniques can be employed to extract information about herbal properties, traditional uses, and clinical applications from TCM literature. Machine learning models can then be trained on this data to predict the potential efficacy of different herbs and formulas for various diseases. Second, AI can be used to screen large libraries of compounds, including those derived from TCM herbs, to identify molecules that interact with specific drug targets. Virtual screening and molecular docking simulations, powered by AI, can significantly accelerate this process. Third, AI can help elucidate the complex mechanisms of action of TCM formulas by analyzing the interactions between different herbal components and their targets. Network pharmacology approaches, combined with AI, can be used to map out the intricate relationships within TCM formulas and predict their therapeutic effects. Finally, AI can be used to repurpose existing drugs for new TCM-related [11-14] indications. By analyzing the similarities between the mechanisms of action of existing drugs and TCM formulas, AI can identify potential candidates for drug repurposing, significantly reducing the time and cost associated with drug development.
This paper explores the various applications of AI in TCM drug discovery, examining how different AI techniques are being utilized at each stage of the process. We discuss the potential of AI to accelerate target identification, compound screening, mechanism of action prediction, and drug repurposing. We also delve into the challenges and opportunities associated with AI-driven TCM drug discovery, including data standardization, model validation, and the integration of TCM knowledge with modern drug discovery pipelines. By addressing these challenges and harnessing the power of AI, we can unlock the vast potential of TCM as a source of novel therapeutics, offering new hope for treating a wide range of diseases.
AI-driven drug discovery from Traditional Chinese Medicine (TCM), while holding immense promise, faces several significant challenges:
Data acquisition and standardization
• Data scarcity and quality: While TCM possesses a wealth of historical knowledge, much of it is documented in classical Chinese and lacks the structured format required for AI analysis. Modern TCM data, while more accessible, can still be fragmented, inconsistent, and lack standardized terminology. High-quality, well-annotated datasets are crucial for training effective AI models, and acquiring such data remains a major hurdle.
• Data heterogeneity: TCM data comes in various forms, including herbal texts, formulas, clinical records, and experimental data. Integrating these heterogeneous data sources into a unified framework for AI analysis is a complex task. Standardizing data formats and developing ontologies for TCM concepts are essential.
• Data bias: Bias in TCM data, whether due to geographical variations in herbal sources, differences in diagnostic practices, or limited representation of certain populations, can affect the performance and generalizability of AI models. Addressing data bias is crucial for developing robust and reliable AI tools.
Model development and validation
• Feature engineering: Identifying and extracting relevant features from TCM data for AI model training is a challenging task. This requires a deep understanding of TCM principles and expertise in feature engineering techniques. Defining the most informative features that capture the essence of TCM concepts is crucial for building accurate models.
• Model interpretability: “Black box” AI models, while potentially accurate, can be difficult to interpret, making it challenging to understand how they arrive at a particular prediction. Understanding the reasoning behind AI-driven drug discovery is crucial for building trust among researchers and facilitating the translation of AI findings into practical applications.
• Model validation: Rigorous validation is necessary to ensure the accuracy and reliability of AI models. This involves testing the models on diverse datasets and comparing their performance against existing methods. Validating AI [15-18] models in real-world experimental settings is crucial for assessing their practical utility.
• Integrating TCM knowledge: Effectively integrating the vast body of TCM knowledge, including herbal properties, formulas, and traditional uses, into AI models is a significant challenge. Developing methods to represent and reason with TCM knowledge within AI frameworks is crucial for building more informed and accurate models.
Biological complexity and mechanism of action
• Multi-target interactions: TCM formulas often contain multiple herbs that act on multiple targets simultaneously. Understanding the complex interactions between different herbal components and their targets is crucial for elucidating the mechanisms of action of TCM formulas. AI models need to be able to capture these complex interactions.
• Synergistic effects: The synergistic effects of different herbs within a TCM formula can be difficult to predict using traditional drug discovery approaches. AI can potentially help identify and understand these synergistic effects, but this requires advanced modeling techniques.
• Mechanism of action elucidation: While AI can help identify potential targets and pathways, elucidating the precise mechanisms of action of TCM components remains a significant challenge. Further experimental validation is necessary to confirm AI predictions and understand the underlying biology.
Ethical and regulatory considerations
• Intellectual property: Protecting intellectual property related to AI-discovered TCM drugs is crucial for incentivizing research and development. Clear guidelines and regulations are needed to address intellectual property issues in this area.
• Regulatory pathways: Navigating the regulatory pathways for AI-driven TCM drug discovery can be challenging. Clear regulatory frameworks are needed to ensure the safety and efficacy of AI-discovered TCM therapeutics.
• Transparency and trust: Building trust in AI-driven TCM drug discovery requires transparency in how AI models are developed and used. Addressing potential biases and ensuring data privacy are crucial for building public trust.
Bridging the gap between TCM and modern science
• Translating TCM concepts: TCM concepts, often described using traditional Chinese terminology, can be difficult to translate into a format that AI can understand. Bridging the gap between TCM’s traditional language and modern scientific concepts is a significant challenge.
• Integrating TCM with modern drug discovery: Integrating AI-powered [19-22] TCM drug discovery with modern drug discovery pipelines requires collaboration between TCM practitioners, AI researchers, and pharmaceutical companies. Building bridges between these different disciplines is essential.
AI-driven drug discovery from Traditional Chinese Medicine (TCM) offers a multitude of potential benefits, significantly accelerating the development of novel therapeutics:
Accelerated drug discovery process
• Faster target identification: AI algorithms can analyze vast amounts of TCM data, including herbal texts, formulas, and biological data, to identify potential drug targets much faster than traditional methods. This speeds up the initial stages of drug discovery.
• Efficient compound screening: AI-powered virtual screening and molecular docking can rapidly evaluate the interactions between a large number of compounds and target molecules, prioritizing those with the highest potential for further investigation. This greatly accelerates the compound screening process.
• Reduced development time: By streamlining target identification, compound screening, and mechanism of action prediction, AI can significantly reduce the overall time required to develop new drugs from TCM. This means potentially faster access to new therapies for patients.
Improved efficiency and cost-effectiveness
• Reduced research and development costs: AI can minimize the need for costly and time-consuming laboratory experiments by prioritizing promising candidates for further investigation. This reduces the overall cost of drug development.
• Optimized resource allocation: AI can help researchers allocate resources more efficiently by focusing on the most promising drug candidates and avoiding dead ends. This optimizes the use of research funding and personnel.
• Increased success rate: By improving target identification and compound screening, AI can increase the success rate of drug development, leading to a higher number of approved drugs.
Enhanced understanding of TCM mechanisms
• Mechanism of action elucidation: AI can help decipher the complex mechanisms of action of TCM formulas by analyzing the interactions between different herbal components and their targets. This can lead to a deeper understanding of how TCM works and facilitate the development of more targeted therapies.
• Synergistic effect prediction: AI can potentially identify and understand the synergistic effects of different herbs within a TCM formula, which are often difficult to predict using traditional methods. This can lead to the development of more effective combination therapies.
• Network pharmacology insights: AI-powered network pharmacology approaches can map out the intricate relationships within TCM formulas and predict their therapeutic effects, providing valuable insights into the holistic nature of TCM.
Novel therapeutic opportunities
• Identification of novel drug targets: AI can help identify novel drug targets that may not have been considered using traditional methods. This can open up new avenues for treating a wide range of diseases.
• Discovery of new active compounds: AI [23-25] can facilitate the discovery of new active compounds from TCM herbs, potentially leading to the development of innovative therapeutics for unmet medical needs.
• Drug repurposing: AI can be used to repurpose existing drugs for new TCM-related indications, significantly reducing the time and cost associated with drug development.
Personalized medicine
• Tailored therapies: AI can analyze individual patient data, including genetic information and TCM diagnostic information, to tailor treatment plans based on their unique constitution and presentation. This aligns with TCM’s emphasis on personalized medicine.
• Predictive diagnostics: AI can potentially be used to develop predictive diagnostic tools that can identify individuals at risk for certain diseases, allowing for earlier intervention and more effective treatment.
Integration of TCM with modern medicine
• Bridging the gap: AI can help bridge the gap between TCM and modern medicine by translating TCM concepts into a format that can be understood by modern scientists. This can facilitate collaboration between TCM practitioners and conventional medical professionals.
• Evidence-based TCM: By providing data-driven insights into the mechanisms of action of TCM therapies, AI can contribute to the growing body of scientific evidence supporting the efficacy of TCM.
The future of AI-driven drug discovery from Traditional Chinese Medicine (TCM) is ripe with possibilities, building upon current advancements and addressing existing challenges. Here are some key directions for future works:
Enhanced data acquisition and integration
• Multi-omics data integration: Integrating multi-omics data (genomics, transcriptomics, proteomics, metabolomics) with TCM data can provide a more comprehensive understanding of the mechanisms of action of TCM formulas and facilitate the identification of novel drug targets.
• Standardized TCM knowledge base: Developing a comprehensive and standardized TCM knowledge base, including herbal properties, formulas, traditional uses, and modern scientific data, is crucial for building robust AI models. This should incorporate ontologies and controlled vocabularies.
• Real-world data integration: Integrating real-world data from clinical practice, such as patient records and treatment outcomes, can provide valuable insights into the effectiveness of TCM therapies and inform drug discovery efforts.
• Data sharing and collaboration: Establishing platforms for data sharing and collaboration among researchers is essential for accelerating AI-driven TCM drug discovery. This requires addressing data privacy and security concerns.
Advanced AI model development
• Explainable AI (XAI): Developing more explainable AI [26-28] models is crucial for building trust and acceptance among researchers and facilitating the translation of AI findings into practical applications. XAI can provide insights into how AI models arrive at a particular prediction.
• Graph neural networks: Utilizing graph neural networks to model the complex interactions within TCM formulas and biological systems can provide a more nuanced understanding of their mechanisms of action.
• Deep learning for mechanism elucidation: Employing deep learning techniques to analyze multi-omics data and identify key pathways and targets involved in the therapeutic effects of TCM formulas.
• Reinforcement learning for drug design: Exploring the use of reinforcement learning for designing novel TCM-inspired molecules with desired properties.
Enhanced understanding of TCM principles
• Integrating TCM theory: Developing methods to effectively integrate TCM theory and principles into AI models is crucial for building more informed and accurate models. This requires translating TCM concepts into a format that can be understood by AI.
• Personalized TCM drug discovery: Developing AI models that can analyze individual patient data, including genetic information and TCM diagnostic information, to tailor drug discovery efforts based on their unique constitution and presentation.
• Synergistic effect prediction: Developing AI models that can predict and understand the synergistic effects of different herbs within a TCM formula, which are often difficult to predict using traditional methods.
Experimental validation and clinical translation
• High-throughput screening: Integrating AI-powered target identification and compound screening with high-throughput experimental techniques can accelerate the validation of AI predictions.
• Mechanism of action studies: Conducting detailed experimental studies to validate the predicted mechanisms of action of TCM components and formulas.
• Clinical trials: Conducting well-designed clinical trials to evaluate the safety and efficacy of AI-discovered TCM drugs.
Ethical and regulatory considerations
• Developing ethical guidelines: Establishing clear ethical guidelines for the development and deployment of AI in TCM drug discovery is crucial for ensuring responsible and ethical use of this technology.
• Regulatory frameworks: Developing clear regulatory frameworks for AI-driven TCM drug discovery is essential for ensuring the safety and efficacy of AI-discovered TCM therapeutics.
• Transparency and trust: Building trust in AI-driven [29-31] TCM drug discovery requires transparency in how AI models are developed and used. Addressing potential biases and ensuring data privacy are crucial for building public trust.
Interdisciplinary collaboration
• Bridging the gap: Fostering interdisciplinary collaboration among TCM practitioners, AI researchers, chemists, biologists, pharmacologists, and clinicians is essential for realizing the full potential of AI in TCM drug discovery.
In conclusion, the application of Artificial Intelligence (AI) [32] to drug discovery from Traditional Chinese Medicine (TCM) represents a paradigm shift with the potential to revolutionize the development of novel therapeutics. TCM, with its rich history and vast library of natural products, offers a unique and valuable resource for modern drug discovery. However, the complexity of TCM formulas, the lack of systematic investigation, and the sheer volume of TCM literature have historically hindered the efficient translation of TCM knowledge into modern drug discovery pipelines. AI offers a powerful solution to these challenges.
This paper has explored the various ways in which AI can be leveraged at each stage of TCM drug discovery, from target identification and compound screening to mechanism of action prediction and drug repurposing. We have discussed the potential of AI to accelerate the drug discovery process, improve efficiency and cost-effectiveness, enhance our understanding of TCM mechanisms, and unlock novel therapeutic opportunities. AI can help decipher the complex interactions within TCM formulas, predict synergistic effects, and identify novel drug targets that may have been overlooked using traditional methods. Furthermore, AI can facilitate the integration of TCM with modern medicine, bridging the gap between ancient wisdom and cutting-edge technology.