Advancing algorithmic drug product development: Recommendations for machine learning approaches in drug formulation.

Area of Science:

• Pharmaceutical Science
• Computational Chemistry
• Biotechnology

Background:

• Artificial intelligence (AI) offers transformative potential in the pharmaceutical industry, impacting drug discovery, development, and clinical practice.
• Machine learning (ML), a subset of AI, has significantly advanced in silico modeling and clinical translation.
• Current data-driven modeling in drug formulation development faces challenges, including limited specific guidance and suboptimal practices leading to unreliable predictions.

Conclusions:

• Current machine learning practices in drug formulation may yield unreliable predictions due to a lack of transparency and interpretability.
• Recommendations are presented to enhance the trustworthiness, transparency, and reliability of machine learning models in pharmaceutical development.
• Future research should focus on developing robust models that offer practical guidance to formulators, moving beyond black-box approaches.