Objective: COVID-19, caused by SARS-CoV-2, highlighted the need for the development of therapeutic strategies. Since drug discovery processes are time-consuming and costly, the use of bioinformatics approaches and molecular modeling has gained considerable attention. In this context, therapeutic peptides, due to their high specificity and flexibility, are considered promising candidates for the development of antiviral drugs.

Method: Inhibitory peptides targeting the SARS-CoV-2 main protease (Mpro) were designed and evaluated. Molecular docking was performed using AutoDock, and optimized grid parameters were used in PyRx for virtual screening. Peptides of 2–10 amino acids were generated using unidirectional and bidirectional incremental construction approaches. After structural optimization, peptides were evaluated with a reference ligand and analyzed for binding interactions.

Results: Bidirectionally designed peptides showed more favorable binding energies and better performance than unidirectional ones. Increasing peptide length generally improved binding affinity and complex stability. Decapeptides exhibited the strongest interactions. Several peptides showed lower binding energies than the crystallographic ligand and RMSD values close to zero, indicating stable complexes.

Conclusions: Longer peptides, especially those designed via bidirectional construction, enhance binding to Mpro and improve inhibitory potential. These results provide a rational basis for designing novel peptide-based inhibitors against SARS-CoV-2.