Natural products(NPs)derived from plants,microbes,and marine organisms have historically been a cornerstone of pharmacotherapy,accounting for approximately 35%of FDA-approved small-molecule drugs since 19811.These com...Natural products(NPs)derived from plants,microbes,and marine organisms have historically been a cornerstone of pharmacotherapy,accounting for approximately 35%of FDA-approved small-molecule drugs since 19811.These compounds have been instrumental in drug discovery,particularly for cancer and infectious diseases,as well as in other therapeutic areas such as cardiovascular diseases(e.g.,statins)and multiple sclerosis(e.g.,fingolimod).展开更多
To the Editor:The recently published correspondence by Farhat[1]is a timely and welcome contribution to the intersection of artificial intelligence(AI)and biomedical research.For the interdisciplinary field of AI and ...To the Editor:The recently published correspondence by Farhat[1]is a timely and welcome contribution to the intersection of artificial intelligence(AI)and biomedical research.For the interdisciplinary field of AI and health-care,I found the article encouraging and thought-provoking.The author admirably emphasizes the potential of DeepSeek-R1 to transform drug discovery,particularly by highlighting its unique mix-ture of experts and chain of thought frameworks that reportedly outper-form proprietary models such as ChatGPT-4o in tasks such as protein-ligand interaction prediction.These insights open a fruitful discourse on domain-specific optimization in large language models for biomedical applications.展开更多
摘要Natural products(NPs)derived from plants,microbes,and marine organisms have historically been a cornerstone of pharmacotherapy,accounting for approximately 35%of FDA-approved small-molecule drugs since 19811.These compounds have been instrumental in drug discovery,particularly for cancer and infectious diseases,as well as in other therapeutic areas such as cardiovascular diseases(e.g.,statins)and multiple sclerosis(e.g.,fingolimod).
摘要To the Editor:The recently published correspondence by Farhat[1]is a timely and welcome contribution to the intersection of artificial intelligence(AI)and biomedical research.For the interdisciplinary field of AI and health-care,I found the article encouraging and thought-provoking.The author admirably emphasizes the potential of DeepSeek-R1 to transform drug discovery,particularly by highlighting its unique mix-ture of experts and chain of thought frameworks that reportedly outper-form proprietary models such as ChatGPT-4o in tasks such as protein-ligand interaction prediction.These insights open a fruitful discourse on domain-specific optimization in large language models for biomedical applications.