Special Session 3 - Combating Communicable and Non-communicable Diseases with Advanced Health Technology

Computation has become a cornerstone of nearly all scientific disciplines, notably in biology and medicine. In light of the extensive data available, computation transcends its traditional role as a mere tool for data processing; it has emerged as an indispensable instrument for uncovering and articulating the scientific narratives embedded within the data. The crux of employing computation to propel scientific discovery lies in the effective translation of applied problems into computationally solvable frameworks, which represents a primary challenge within the field of computational systems biology. In this talk, I will present insights derived from two lines of ongoing research at the HK JC STEM Laboratory of Genomics & AI in Healthcare at PolyU. Firstly, I will discuss the identification of polygenic markers linked to complex diseases, particularly psychiatric disorders, including depression and schizophrenia. This research constitutes a critical aspect of the recently funded theme-based Strategic Target Grant (STG) project, which aims to develop innovative artificial intelligence and genomic technologies to improve mental healthcare services for major psychiatric disorders. Secondly, I will examine the discovery of novel classes of circular RNA and its clinical implications for cancer diagnosis and treatment. The principal takeaway from this discussion highlights the growing integration of computational methodologies, including the latest advancements in AI techniques, at the forefront of scientific research-a trend that warrants our utmost attention.

The integration of high-throughput sequencing technologies has revolutionized pathogen detection and identification across human, animal, and environmental health sectors, exemplifying the One Health approach. This presentation explores how advanced genomic tools, including whole-genome sequencing, metagenomic shotgun sequencing (mNGS), and target capture sequencing, address emerging pathogens in our rapidly changing global environment. These technologies enable unprecedented capabilities for pathogen identification, antimicrobial resistance tracking, and cross-species disease transmission monitoring.


As climate change and urbanization accelerate the emergence of novel pathogens, the One Health framework underscores the necessity for interdisciplinary collaboration and standardized methodologies to strengthen global health security. We will present our real-world implementation of these genomic tools across three interconnected domains:


•    Clinical settings: Diagnosing acute invasive diseases and investigating nosocomial outbreaks
•    Environmental settings: Conducting wastewater surveillance and investigating urban melioidosis cases
•    Animal and food safety: Studying melioidosis in non-human primates and surveilling foodborne multidrug-resistant organisms


Our findings demonstrate how these technologies create critical bridges between clinicians, veterinarians, and environmental engineers, significantly enhancing our collective ability to detect, monitor, and respond to emerging infectious threats. By fostering collaborative networks and implementing standardized practices across disciplines, these innovative genomic approaches are building a more resilient and integrated global health system capable of addressing complex challenges at the human-animal-environment interface.

Cancer is a complex group of multi-omics disorders. Integrated multi-omics analysis enables comprehensive evaluation of tumor genomic, biochemical, metabolic, proteomic, and epigenomic processes that drive tumor development. These alterations are crucial for precision cancer diagnosis, risk stratification, treatment selection, and prevention. We have implemented this integrated approach that leverages advanced bioinformatics and Artificial Intelligence based workflows for multi-omics analysis of over 10,000 tumors including more than 6,000 pediatric tumors and provided comprehensive tumor multi-omics profiles to inform precision patient care. Over 90% of the patients have received clinically impactful results, including 82% with diagnostic significance leading to changing or modifying diagnosis for more than 10% of the patients based on the multi-omics data. Additionally, 50% of patients received prognostic insights and 44% benefited from therapeutic guidance, including approximately 15% with biomarker-based targeted therapies. Furthermore, confirmed or presumed pathogenic/likely pathogenic germline variants were identified in 5% to 10% of hematological malignancies and up to 20% of pediatric patients with brain tumors. As our knowledge about tumor multi-omics expands and the inclusion of more and more biomarkers into the professional guidelines, the percentages of patients receiving updated diagnoses and risk-based or biomarker-based targeted therapies upsurge steadily. 
This integrated multi-omics approach is especially beneficial to patients with hard-to-diagnose tumors and/or difficult-to-threat, i.e., refractory/recurrent, tumors. The identification of associated cancer predisposition in pediatric cancer patients also facilitates cascade genetic testing for at-risk family members, cancer surveillance, and early cancer detection for mutation carriers. Given the significant success of our program, we advocate for the widespread application of this multi-omics approach in both pediatric and adult cancer care, not only to cure cancer but also to preserve patients' quality of life.