Translating “AI for omics” into precision therapy

Abstract

Precision medicine aims to empower clinicians to predict the most appropriate course of action for patients with complex diseases like cancer and others.1 With an efficient interrogation of the omics, molecular, and clinical data at play in diseases, effective, personalized, and precise medical treatment strategies are expected for many disorders. In addition, the treatment modality of precision medicine is increasingly diversified, spanning from mainstream drug and antibody treatment to newly developed gene and cell therapy, etc. In various treatment modalities including the treatment by Chinese medicine, precision is the highest demand, and the underlying patterns and rationales across various factors to uncover the biological mechanism and actionable information that support early detection, prevention and therapy of complex disorders can be identified by the omics data.1,2 Therefore, the omics sequencing technology and accumulated omics data serve as the core technology and resources that can have huge impact on such a course of action. Accordingly, the integrative analysis of such omics data provides a great opportunity to support precision medicine studies. However, facing the unique features of the omics data, like the dominant characteristics of “high dimension and small sample”, traditional data analysis strategies are likely unsuitable, and data-driven artificial intelligence (AI) technology is emerging as an effective paradigm for precision medicine study. To this end, this study aims to present a concise and overview perspective of the characteristics and the emerging directions of omics sequencing data, as well as the related AI analysis schema in handling such data, together with the summarized precision therapy cases based on the integration of AI and omics.