Whole Genome Sequencing-based digital twin as a clinical decision support system: in-born disease risks and pharmacogenomics with Large Language Model

Abstract

Whole Genome Sequencing (WGS) utilizing Next Generation Sequencing has emerged as a groundbreaking tool in modern medicine, providing unparalleled insights into an individual's genetic profile. This comprehensive genetic analysis has the potential to revolutionize clinical decision-making by equipping clinicians with critical information crucial for diagnosis, treatment selection, and patient management. It holds a particular promise in identifying rare and complex genetic conditions, pharmacogenomics, and facilitating differential diagnoses. However, the utilization of a PDF-based report has raised concerns related to false positives, primarily attributable to limited space constraints that necessitate a focused selection of variants and phenotypes.Here we introduce a DNA-based digital twin hosted in a cloud-based platform, offering unrestricted access to complete variant information. To streamline clinical decision-making, we have categorized in-born disease risks and pharmacogenomics into four tiers. Integration with a Large Language Model such as ChatGPT empowers clinicians to perform differential diagnoses based on symptoms and genetic susceptibility. As the cost of whole genome sequencing continues to decrease, its in-clinic adoption becomes increasingly accessible and cost-effective. The seamless integration of WGS into routine clinical practice holds the potential to significantly augment diagnostic precision, treatment effectiveness, and overall healthcare quality.