[Department of Ocean Science]: MPhil Thesis Defense Seminar: DNA Methylation-Based Age Estimation for Cetaceans in Hong Kong waters Using Methylation-Sensitive High-Resolution Melting Analysis
 

Abstract: 

This thesis developed and validated a DNA methylation-based age assessment framework for cetaceans in Hong Kong waters using methylation-sensitive high-resolution melting (MS-HRM) analysis. First, the MS-HRM workflow was optimized by comparing two approaches for methylation standard-curve construction: tissue-derived and de novo synthesized standard DNA. Also, five regression models were evaluated for the standard curve using the Akaike Information Criterion (AIC). Both standard DNA approaches generated reliable standard curves, while de novo synthesized standard DNA provided a more reproducible workflow. Using the optimized workflow, DNA methylation-based age estimation models were established for the Indo-Pacific finless porpoise (Neophocaena phocaenoides). Methylation rates of three age-associated loci, GRIA2, CDKN2A, and TET2, were quantified using MS-HRM, and support vector regression (SVR) was used to integrate methylation signals to estimate chronological age. Although the individual markers showed weak linear associations with chronological age, the SVR model using the three genes achieved good prediction performance, with a mean absolute error of 1.35 years, a root mean square error of 1.81 years, and an R² of 0.620. Furthermore, the applicability of the optimized MS-HRM framework was evaluated in other cetaceans. A preliminary analysis of Indo-Pacific humpback dolphins (Sousa chinensis) was conducted, although a limited sample size reduced model performance. In addition, blood-derived DNA from Indo-Pacific bottlenose dolphins (Tursiops aduncus) at Ocean Park Hong Kong was successfully analyzed, demonstrating that the workflow could be applied to minimally invasive blood samples from living cetaceans. Overall, this study demonstrates that MS-HRM provides a rapid and practical approach for cetacean DNA methylation analysis and age estimation, with potential applications in demographic monitoring and conservation management.