Abstract: To solve the problem of poor thermal stability of the current D-allulose 3-epimerase (DAEase), the ancestor sequences of DAEase with different catalytic domains were reconstructed by big data mining, reasonable modification and ancestor sequence reconstruction (ASR) strategy under the guidance of phylogenetic information. The expression vectors of the ancestor sequences were constructed, and DAEase A13 with significantly enhanced thermal stability was screened by recombinant expression and molecular docking, and its enzymatic properties were characterized. In addition, the molecular mechanism of thermal stability enhancement of DAEase A13 was revealed based on structural analysis and molecular dynamics. The results showed that the half-life of A13 constructed based on ASR strategy could reach 8.4 h at 70 ℃, indicating that its thermal stability was significantly enhanced compared with that of wild-type (WT) enzyme. The maximum conversion rate of A13 reached 31%, indicating that the catalytic activity of A13 was slightly higher than that of WT enzyme. The structural and molecular dynamics analysis revealed that the increase in hydrogen bonding and hydrophobic interaction in ASR A13 was the main factor responsible for maintaining the stability of the enzyme's molecular structure at high temperatures. The results showed that ASR strategy could modify DAEases to enhance the stability, activity or hybridity, which could provide superior biocatalyst sources for various industrial applications of functional sugars.