An Efficient Drug-Drug Interactions Prediction Technology for Molecularly Intelligent Manufacturing

Drug-Drug interactions (DDIs) are considered as the unwanted side, causing patients irreparable side effects [1] due to the concurrent consumption of two or more drugs [2-4]. As a fundamental research issue in the field of molecular biology, DDIs has received great attention. The resources of molecular biology, such as research literature, structure diagrams, and interactive genes analyses, have the substantial increments [5-7]. Researchers are facing enormous amount of materials to review, which requires considerable energy to investigate and organize crucial information with difficulty. The achievements of Natural Language Processing (NLP) could accelerate this procedure and improve the reliability for their further research, in the meantime, the linguistics could take advantages of the large, well-curated resources as well [5], benefiting the both groups. Considering the importance of DDIs in humans health, industry and the economy, and the substantial amount of cost and time of experimental approaches [1, 8], it is appropriate to introduce the automatic methodologies into this field. In the work by Zhang et al. [9], they argue that Information Extraction (IE) applications, such as discovering DDIs, must move beyond a scenario where relevant information is coherently and explicitly stated within a single document.