Dengue hemorrhagic fever (DHF) is life-threatening severe dengue with a hallmark of vascular leakage. A mast cell protease, beta-tryptase, has been found to promote vascular leakage in DHF patients, which could be a potential target for the treatment of DHF. This study aims to develop a theoretical background for the design and selection of beta-tryptase inhibitors through these approaches: two-dimensional quantitative structure-activity relationships (2D-QSAR) study, molecular docking analysis, molecular dynamics (MD) simulation, and structure-based pharmacophore modelling (PM). A total of 34 beta-keto-[1,2,3]-oxadiazoles scaffold-based compounds, obtained from a past study, were used to generate 2D-QSAR models by Genetic Function Approximation (GFA). The generated 2D-QSAR models were used to investigate the relationships between the molecular structure and the potency of beta-tryptase inhibition. Molecular docking explores the binding affinities and binding interactions of the beta-keto-[1,2,3]-oxadiazoles scaffold-based compounds with beta-tryptase (PDB Code 4A6L) by the CDOCKER tool in Discovery Studio. In addition, MD simulation was performed using GROMACS on the docked complex of the reported most active compound, compound 11e, to study the binding mechanism of the compound towards beta-tryptase. Finally, a structure-based pharmacophore model was generated from the same docked complex to identify the important features that contribute positively to the inhibitory activity of the compound towards beta-tryptase. The best 2D-QSAR model has demonstrated statistically significant results through its r2, q2, and r2 (pred) values of 0.9077, 0.733, and 0.8104, respectively. The docking results of compound 11e showed lower CDOCKER energy than the 4A6L co-crystallised ligand, indicating good binding affinity. Furthermore, compound 11e has a similar binding pattern as the 4A6L co-crystallised ligand, which involves the binding of active residues such as Asp207, SER208, and GLY237. The MD simulation shows that the 4A6L-compound 11e complex has RMSD below 2 angstrom throughout the 100ns of simulation, indicating that the docked complex is stable. Besides, MD simulation showed that the inhibitory potency of compound 11e is contributed by hydrogen bonding with 4A6L active site residues, which are ASP207, SER208, and GLY237. The best pharmacophore model identified features that contribute to the inhibitory potency of a compound, which included hydrogen bond acceptor, ionic interaction, hydrophobic interaction, and aromatic ring. This study has fundamentally supplied valuable insight and knowledge for developing novel chemical compounds with improved inhibitory ability against human beta-tryptase.