Establishment of a simple prediction method for DNA melting temperature: high-resolution melting curve analysis of PCR products

High-resolution melting analysis is a technique that leverages the principle that the thermal stability of dsDNA is influenced by its length and base composition. This method generates a melting curve by real-time monitoring of the changes in fluorescence signal as dsDNA melts during the heating process. The melting temperature serves as a fundamental indicator of sample characteristics in HRM analysis. During the initial stages of designing a new HRM experimental system, accurately predicting the Tm position of the established system can significantly enhance research efficiency. Currently, there is a limited number of studies focused on the prediction of Tm values in HRM analysis, with varying levels of predictive accuracy. The nearest-neighbor method model can well reflect the interaction of adjacent base pairs. Therefore, we combined the nearest neighbor method model and applied parameters such as enthalpy change, entropy change, GC content and number of base pairs of the DNA sequence to derive a new empirical formula for predicting Tm values. In this study, five species of seawater diatoms were selected as the research subjects. Four specific primers were employed to amplify the extracted DNA through PCR, and the resulting amplified products underwent HRM analysis and Sanger sequencing. Based on the obtained DNA sequences, we calculated the corresponding GC content, number of base pairs, enthalpy change and entropy change, combined with the Tm value obtained from the experiment. Finally, the following formula for predicting Tm value is obtained: (1) When the GC content is 40%≤GC content≤60%: Tm=ΔH/ΔS-0.27GC%-(150+2n)/n-273.15; (2) When the GC content is <40%: Tm=ΔH/ΔS-GC%/3-(150+2n)/n-273.15. After that, the DNA sequences amplified using two other specific primers were verified, and the predicted Tm values were compared with the measured values. The average error was within 1 °C.The results show that the formula obtained in the study can accurately predict the Tm value, which can be effectively used to identify the species of unknown samples.Therefore, this Tm value prediction method provides new methods and ideas for solving practical problems in multiple related fields.