PeptiTools: An in silico Screening Platform for Venom-Derived Peptides Using Cm 38-3 as a Case Study

Animal venom represents a promising source of diverse bioactive molecules, including those with antimicrobial activity. However, the large number of candidates and the costs of synthesis and experimental validation limit their investigation. To address these limitations, our research group developed PeptiTools, an in silico screening platform that integrates physicochemical and structural analyses with exploratory models of membrane interactions, using Monte Carlo sampling and biological targets to support the rational selection and optimization of bioactive peptides. Here, we present PeptiTools, its development, and its application using Cm 38, a peptide isolated from the venom of the scorpion Centruroides margaritatus, as a case study. The screening workflow involved 64 antimicrobial peptides, including activity prediction, molecular characterization, and analysis of interactions with a target protein. We selected Cm 38 for rational optimization after it received a predicted activity score of 38%. Amino acid substitutions yielded the Cm 38-3 variant, with a predicted activity score of 99%. In molecular docking against DNA gyrase A, the HADDOCK score improved from −78.9 ± 8.2 for the native peptide to −87.8 ± 4.6 for the mutant, with hydrogen bonds increasing from three to four. During 600 ns molecular dynamics simulations, Cm 38-3 maintained a helical conformation, whereas the native sequence was more disordered. Microbiological assays showed a minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) of 1.68 µM against Klebsiella pneumoniae for Cm-38-3, whereas the native peptide showed no activity at concentrations up to 50 µM under the conditions tested. We also observed MICs of 8 µM against Staphylococcus aureus and 4 µM against Staphylococcus haemolyticus, as well as an MIC and MBC of 3.125 µM against Escherichia coli.  Overall, this study highlights both our research group's development of PeptiTools as an integrated computational platform and its practical use in prioritizing and rationally optimizing venom-derived antimicrobial peptides. By combining computational predictions with experimental testing, PeptiTools offers a systematic way to accelerate the identification of promising peptide candidates while reducing unnecessary synthesis and experimental screening.