Abstract: In recent years, the deep integration of artificial intelligence (AI) with pharmaceutical sciences has offered new opportunities to overcome the traditional bottlenecks of drug development characterized by high costs and low efficiency. Owing to their wide coverage and rich information content, biological data have been widely applied across various stages of AI-driven drug discovery. This paper summarizes the main types and scales of biological and pharmaceutical data, including multi-omics data, protein-protein interaction networks, chemical structure-bioactivity correlation data, and three-dimensional structural information, and discusses recent advances in the application of these data to key stages of drug development, such as target identification, lead compound discovery and optimization, and druggability evaluation, aiming to provide some reference for constructing a standardized, multi-modally integrated and clinically translatable biodatadriven drug design system.