Computational design of peptide ligands that can potently and specifically recognize and bind to disease-related protein targets has attracted great interest in the biomedicinal community.However,peptide is highly flexible ligand that would incur large entropy penalty upon binding to protein receptor,and conformational restriction strategy is thus used naturally and artificially to improve peptide rigidity and to pre-structure peptide ligand in active conformation that can considerably facilitate the protein binding.Here,we highlight proline-mediated conformational restriction which present as the simplest strategy in the natural world and has been shown to enhance their binding affinity/specificity and metabolic stability by different numbers and distribution patterns of proline in peptide binding protein.We utilize several tools like pymol and VMD to design mutations of functional proline in peptide based on protein domains (SH3,GYF,EVH1,profilin,and WW) that have known structures and interact with proline-rich peptides.Then molecular dynamics (MD) simulation is used to discover the structural comparison and computational binding affinity prediction to investigate entropy and enthalpy change between each mutation system and natural system.Results indicate that proline-mediated conformational restriction effect plays great role in protein-peptide recognition.Moreover we propose that this study could have implications for the future development of peptides designed.