抗菌肽结构改造与人工智能研发策略

Antimicrobial peptides: structure modification and development with artificial intelligence

摘要:

抗菌肽是一类广泛存在于生物体内的小分子肽，参与构成生物体先天免疫，可以有效抵抗病原微生物的入侵。抗菌肽具有广谱抗菌活性，且不易产生耐药性等特点，在治疗感染性疾病方面具有独特的优势，有望成为理想的抗感染药物。然而，由于部分抗菌肽尚存在稳定性差、毒性高等问题，限制了抗菌肽的广泛应用。由于人工智能算法能有效合成具有高稳定性、低毒性的抗菌肽，在探索天然抗菌肽中展现了巨大的潜力，因此本文简述了抗菌肽的抗菌机制、结构改造以及利用机器学习和深度学习等人工智能算法进行新型抗菌肽研发的优化策略，以期为抗菌肽结构优化及研发提供新思路。

Abstract:

Antimicrobial peptides are peptides with low molecular weight found in almost all forms of life. They are part of the innate immune response of all classes of life, having broad-spectrum antimicrobial activity and low potential to elicit resistance. Thus, they have unique advantages in combating infections and demonstrate the potential as ideal anti-infective agents. However, some problems such as poor stability and high toxicity limit their application. In recent years, it has been found that artificial intelligence can help develop stable antimicrobial peptides with low toxicity, showing great potential in exploring natural antimicrobial peptides. In this review, we briefly summarized the antimicrobial mechanism and structure modification of antimicrobial peptides as well as the strategy of using artificial intelligence algorithms such as machine learning and deep learning for research and development of antimicrobial peptides. This review is expected to provide new mindset for the structure optimization and development of antimicrobial peptides.