EvoPepFold: A Hybrid Evolutionary and Structural Pipeline for AI- Guided Peptide Inhibitor Design Using AlphaFold and Rosetta

EvoPepFold: A Hybrid Evolutionary and Structural Pipeline for AI- Guided Peptide Inhibitor Design Using AlphaFold and Rosetta | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article EvoPepFold: A Hybrid Evolutionary and Structural Pipeline for AI- Guided Peptide Inhibitor Design Using AlphaFold and Rosetta Frederico Chaves, Diego Mariano, Luana Bastos, Ana Paula Abreu, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7706745/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 29 Nov, 2025 Read the published version in Scientific Reports → Version 1 posted 14 You are reading this latest preprint version Abstract Peptide inhibitors represent a promising class of antiviral therapeutics, offering several advantages over traditional small-molecule drugs, including low toxicity, high specificity, and biocompatibility. However, rational and efficient design and optimization of inhibitor peptides remains a significant challenge to current methods. Here we show EvoPepFold, a genetic algorithm-based framework designed to generate inhibitory peptides. We evaluated EvoPepFold to design and optimize peptides targeting the SARS-CoV-2 main protease (M pro ). EvoPepFold was applied through two complementary strategies: molecular docking using the Rosetta suite, and peptide 3D modeling with ColabFold. The top candidates were further evaluated through molecular dynamics simulations to assess stability and interaction energy. Our results demonstrate that EvoPepFold successfully identified peptides with favorable binding affinities and stable protein-peptide interactions. These findings highlight the potential of evolutionary algorithms in guiding the rational design of peptide-based antivirals, contributing to ongoing efforts in peptide engineering for therapeutic applications. Biological sciences/Biochemistry Biological sciences/Computational biology and bioinformatics Biological sciences/Drug discovery Biological sciences/Structural biology protein-peptide interactions COVID-19 Mpro Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Introduction Peptides are molecules composed of short chains of amino acids linked by peptide bonds, typically consisting of 2–50 residues [ 1 ]. In living beings, they play diverse roles in biological systems, acting as hormones, signaling molecules, neurotransmitters, and regulators of various physiological processes [ 2 ]. For example, in the case of infectious diseases, peptides can function as antiviral agents by interfering with key stages of the viral life cycle, such as entry, replication, or assembly. Additionally, peptide inhibitors represent a promising class of antiviral therapeutics, offering several advantages over traditional small-molecule drugs [ 3 ]. Advances in synthetic approaches have enabled the modification of the biophysical and biochemical properties of peptides, making them promising candidates for drug development, mainly due to their low toxicity, high specificity, and biocompatibility [ 4 ]. Currently, more than 60 peptide-based drugs have been approved, with many others in clinical trials [ 5 ]. These compounds have demonstrated effectiveness in treating diseases such as cancer, type 2 diabetes, and autoimmune disorders, with exenatide derivatives standing out as notable examples [ 2 ]. Furthermore, understanding protein–peptide interactions is essential for the rational design of new compounds with therapeutic and biotechnological potential [ 6 ]. Antiviral peptides can be strategically developed to bind to vital viral enzymes or structural components, thereby blocking their activity. As a result, there is growing interest in the pharmaceutical industry in creating peptides capable of interfering with essential protein–protein interactions required for viral replication [ 7 ]. These peptides emerged as promising allies in the fight against viral infections, such as COVID-19. Threats to public health, such as the SARS-CoV-2 virus, highlight the need for methodologies to expedite the development of effective antiviral therapeutics [ 8 ]. While vaccination efforts have significantly mitigated the severity of the pandemic, the emergence of new variants and the need for treatments for infected individuals underscore the importance of antiviral drug discovery. In the particular case of SARS-CoV-2, the main protease (Mpro) is a crucial enzyme responsible for processing viral polyproteins. It has emerged as a prime target for therapeutic intervention due to its essential role in viral replication and its distinct substrate specificity compared to human proteases [ 9 ]. The unique preference of Mpro for a glutamine residue at the P1 position of its substrates presents an opportunity to design highly selective inhibitors [ 10 ]. However, the potential for drug resistance arising from mutations in the Mpro sequence highlights the need for innovative approaches to identify novel antiviral agents [ 9 ]. As discussed, peptide inhibitors represent a promising class of antiviral therapeutics, offering several advantages over traditional small-molecule drugs [ 11 ]. Their inherent ability to interact with large protein surfaces with high specificity and potency makes them well-suited for targeting enzymes like Mpro [ 11 ]. However, designing peptides with high affinity to specific protein complexes is not trivial. In this context, computational strategies have been widely employed to aid in designing peptides with enhanced affinity for the receptor. For example, molecular docking methods, such as HDOCK [ 12 , 13 ], HPEPDOCK [ 14 ], and Rosetta [ 15 , 16 ], can help in identifying the binding poses of the peptide to the protein and predicting their binding energy. Although not explicitly designed for this purpose, AI-based structural modeling tools like AlphaFold Multimer [ 17 ] have also been adopted to simulate protein-peptide interactions [ 18 ]. However, these tools do not perform regular molecular docking; instead, they model the entire complex using deep neural network techniques trained on large datasets of protein-protein interactions. Furthermore, various computational methods can be employed to automate and optimize the generation of peptides with therapeutic properties. Genetic algorithms (GAs) consist of one class of such optimization methods. They were developed inspired by evolution, which iteratively evolve a population of candidate solutions to improve their performance according to a defined fitness function. The use of these algorithms for sequence-based peptide optimization is already widely discussed in the literature [ 19 ]; however, to the best of our knowledge, the combination of molecular docking, AI-based modeling, molecular dynamics simulations, and genetic algorithms to optimize peptides has not yet been explored. In this study, we aimed to identify peptides with the potential to bind to Mpro, thereby helping to develop new therapeutic candidates against SARS-CoV-2. To achieve this, we developed and evaluated the feasibility of an innovative computational pipeline composed of three main steps: (i) a genetic algorithm for the generation and optimization of peptide sequences based on evolutionary criteria; (ii) AlphaFold2 [ 20 ], a deep learning-based tool used to predict the three-dimensional structures of the generated peptides; and (iii) the Rosetta software [ 21 , 22 ] using its scoring function [ 23 ] to estimate the binding energy between each peptide and Mpro. The function estimates this energy based on physicochemical criteria, serving as a relative indicator of complex stability and aiding in the selection of the most promising peptides. Lastly, we evaluated the best proposed peptides using molecular dynamics simulations. Material and methods Data collection We collected, from the Protein Data Bank (PDB) [ 24 ], the 3D structure of the COVID-19 main protease (Mpro) in complex with an inhibitor N3 peptide (PDB ID: 6LU7) and defined this structure as the target. N3 is a peptide of six amino acid residues (sequence: “XAVLXX”, where X corresponds to a non-canonical amino acid). Additionally, we collected 2,355 peptides composed of 5 to 30 amino acids from the Propedia database [ 25 , 26 ]. These peptides were used to define the initial population for the genetic algorithm to improve on. Mpro docking site definition We used the binding position of the N3 peptide in the Mpro to define the binding region of interest (Fig. 1 A-B). We selected residues with at least one atom within 5 Å of the N3 peptide, and described this region as the docking site contact interface (Fig. 1 C). Finally, we removed the peptide from the complex (Fig. 1 D) and used the remaining structure as a target for further analyses. The following 24 residues were selected from the Mpro interface of contact with the N3 peptide (Fig. 1 E): T24, T25, T26, L27, H41, M49, F140, L141, N142, G143, S144, C145, H164, M165, E166, L167, P168, H172, D187, R188, Q189, T190, A191, and Q192. Our ultimate goal is to propose a peptide that binds to the Mpro structure with more affinity than the N3 peptide. Therefore, this hypothetical peptide should at least occupy the same binding site as the original peptide. To measure this, we define the residue occupancy (RO) parameter. The RO parameter indicates the percentage of the 24 amino acids of Mpro at least 5Å away from the ligand. By default, the original peptide has RO = 100%. A docked peptide with an RO score equal to 0% is bound to a binding site different from the one targeted in this work. Genetic algorithm overview Figure 2 . Overview of the genetic algorithm. Peptides obtained from the Propedia database were initially docked to the Mpro structure. The top 100 results were selected as the initial population for the genetic algorithm. Two strategies were then employed: a docking-based approach (utilizing the Rosetta suite) and an AI-driven 3D modeling approach (utilizing ColabFold). A fitness function evaluated peptide candidates through a tournament selection scheme, and genetic operations were applied to peptide sequences to generate a new population. Each new set of peptide structures was modeled through docking (Rosetta) or 3D modeling (ColabFold), resulting in a population of 25 structures. These steps were iterated over 100 generations. Finally, the best peptides from each generation were selected based on the lowest docking scores. The evolutionary process began with an initial population of 100 peptides. In each generation, new peptides were produced through two main genetic operators: crossover and mutation . The crossover operator recombined segments of parent peptides using variable crossover lengths, which were randomly chosen for each operation. The mutation operator introduced diversity by applying insertion, deletion, or substitution of amino acid residues at random positions in the peptide sequence. Newly generated peptides were modeled using two strategies: docking protein-peptide using Rosetta [ 21 , 22 ] and AI-modelling using ColabFold, an AlphaFold2-powered tool [ 20 ] (further details are provided in the following subsections). The fitness function used to evaluate each peptide was the docking score; however, the occupancy of the binding site was used to filter out peptides that bind elsewhere. The highest-ranking peptides were selected to propagate the next generation. This iterative evolutionary process was repeated until convergence criteria were met, aiming to identify peptides with enhanced binding potential to Mpro. Initial population To define the initial population, docking was performed between the 2,355 peptides collected from Propedia and the Mpro structure using PyRosetta4 - release 2024.39 [ 21 , 22 ]. Due to the large number of peptides to be tested, we performed low-resolution ab initio docking, generating five poses for each peptide, followed by an energy minimization step. To accelerate computation, docking jobs were run in parallel on 50 CPUs (50 peptides processed concurrently, one peptide per CPU). Peptides that did not achieve at least 30% RO were removed from further consideration. The remaining peptides were ranked by their best docking score across the five poses, and the top candidates were selected for downstream analysis. Parameters definition To define appropriate genetic algorithm parameters, a preliminary tuning run was performed using the top 25 peptides from the initial population. For each tested parameter combination, the GA was executed for 20 generations using a reduced-fidelity setup, where only one docking pose per peptide was computed using Rosetta [ 21 , 22 ]. Although this approach does not adhere to best practices for docking-based scoring, it significantly reduces computational cost and enables the rapid comparison of parameter effectiveness. The goal of this experiment was not to estimate binding affinity, but to evaluate the relative impact of key parameters, including mutation rate, tournament size, crossover rate, and elitism, on optimization performance. The best-performing parameter set from this coarse search was then adopted for the full-scale pipeline, where higher-resolution scoring was applied. Operations The population was subjected to two types of genetic operations: crossover and mutation (Fig. 3 ). In the crossover operation, peptide segments were exchanged between two parent sequences, generating new peptide variants. In the mutation operation, diversity was introduced through random substitution, deletion, or insertion of amino acid residues. The type of operation applied at each step was chosen probabilistically, with 90% chance of performing crossover and a 10% chance of performing mutation. Structural modeling In the next phase, two modeling strategies were employed to evaluate peptide–Mpro binding. In the first approach, each sequence was modeled as a random conformation peptide and docked to the Mpro using Rosetta’s FlexPepDock [ 23 ] high-resolution ab initio protocol, generating ten poses per peptide. In the second approach, AlphaFold2 [ 20 ] (in multimer mode) was used to directly predict the structure of the peptide–Mpro complex. Both experiments were executed on the same server — an AMD Ryzen Threadripper PRO 5995WX 64-Cores processor equipped with an 80GB NVIDIA A100 GPU. For each evaluation, the total time per generation and the time per peptide were recorded. In the AlphaFold2 run, multiple sequence alignments were generated using ColabFold [ 27 ], and structure prediction was performed locally. Due to resource constraints, a 90-minute break was added between generations to avoid GPU saturation. In the Rosetta run, 25 CPU cores were used to perform docking in parallel, allowing batches of 25 peptides to be processed concurrently. Fitness function The top-performing peptides were identified using a tournament-based selection strategy. Complexes generated from both approaches were evaluated with Rosetta’s energy function [ 23 ], which estimates binding free energy at the peptide–protein interface. The PyRosetta library [ 28 ] was used to assess binding energy and automate the score calculations. These scores guided the GA. In each variation, the best-performing peptide in each generation was retained via elitism. At the same time, the remaining population underwent tournament selection, mutation, and crossover to create the next generation of sequences. The best peptide can be maintained for up to three generations. This process was repeated for 100 generations, allowing the algorithm to converge on peptides with progressively improved predicted binding characteristics. Molecular dynamics simulations To evaluate the best results proposed by the case study, we performed molecular dynamics (MD) simulation experiments. The simulations were performed using GROMACS [ 29 ] with the CHARMM36 force field [ 30 , 31 ] and a standard explicit water model on a workstation equipped with CUDA-enabled GPU acceleration (Nvidia A100 80GB). Protein–ligand complexes were placed in a cubic box, centered, solvated, and neutralized with counterions. Energy minimization was performed using the steepest-descent algorithm (50,000 steps). Equilibration proceeded in two stages: NVT for 100 ps at 300 K with a V-rescale thermostat, followed by NPT for 100 ps at 300 K and 1 bar with a V-rescale thermostat and a Parrinello–Rahman barostat, with protein atoms restrained. The production run lasted 100 ns (50,000,000 steps; 2 fs timestep) at 300 K and 1 bar, using the Verlet cutoff scheme and hydrogen-bond constraints. Trajectories were centered and then least-squares-fitted on protein backbone atoms prior to analysis. RMSD (Root Mean Square Deviation) and RMSF (Root Mean Square Fluctuation) were computed with GROMACS Tools and the MDAnalysis Python library [ 32 , 33 ]; plots were generated with Matplotlib [ 34 ]. Binding free energies (ΔG_bind) were estimated by MM-PBSA using gmx_MMPBSA [ 35 ]. Polar solvation energies were obtained by solving the Poisson–Boltzmann equation, and non-polar contributions were estimated from solvent-accessible surface area (SASA). Energies for complex, receptor, and ligand were evaluated over uniformly spaced frames extracted from the production trajectory, and ΔG_bind was computed as ΔG_complex − ΔG_receptor − ΔG_ligand, with ensemble averaging. Per-residue free-energy decomposition was also performed. Results and discussion Peptides proposed to Mpro In the performed experiment, 50 peptides were proposed for each generation of the genetic algorithm. The best peptide is defined based on two metrics: docking energy score of Rosetta (the lower, the better) and occupancy (the higher, the better). Rosetta energies are given on a scale named Rosetta Energy Unit (REU), which is derived from a combination of physics-based and statistical potentials. Figure 4 shows the complexes formed by the binding of Mpro to the peptide that obtained the lowest docking score, and the best peptides from eight generations are shown: G1, G10, G25, G50, G75, G84, G96, and G100. Generations G84 and G96 were chosen because they presented the lowest overall values ​​for the experiments with ColabFold and Rosetta, respectively. The other generations were selected to illustrate how the algorithm evolved peptides. Table 1 presents the top five results for the experiments using ColabFold and Rosetta. Table 1 Top 5 results for ColabFold and Rosetta. REU = Rosetta Energy Unit, which was used as a docking score. ColabFold # Generation Sequence Docking score (REU) Avg. REU Worst REU Time (s) Occupancy 1 84 AGVAKAKAV -634 -525 -96 10,104 79% 2 66 VKAKKCVI -633 -518 -133 8,091 79% 3 59 VAKCCG -628 -509 56 9,019 79% 4 63 AAAKKVTKH -625 -521 -87 7,974 79% 5 48 LAKFKIKH -621 -516 12 44,824 77% Rosetta # Generation Sequence Docking score (REU) Avg. REU Worst REU Time (s) Occupancy 1 96 PGGHSCC -611 -468 448 1,672 70% 2 36 GMLELHQTYT -608 -478 -189 1,765 70% 3 99 GSSSSSYGSGC -606 -495 -456 1,693 67% 4 94 HQSHLSHGCL -605 -486 -428 1,732 70% 5 8 TLILGTERELLESYI -602 -470 -271 1,616 70% For the experiment using ColabFold, the lowest score value was obtained for the peptide AGVAKAKAV obtained in generation 84: -634 REU. In the same generation, the average docking score was − 525 REU, with the worst result being − 96 REU. The occupancy rate was 79% (Table 1 ). For the experiment with Rosetta, the lowest score value was obtained for the peptide PGGHSCC in generation 96 with a docking score of -611 REU and an occupancy of 70%. In the same generation, the average docking score value was − 468 REU, with the worst result being 448 REU. In all evaluated metrics for the top results, the Rosetta results were inferior to those of ColabFold (except for execution time). In this regard, it is important to highlight that the recorded times for Rosetta refer to the generation of 5 poses per peptide, a number far below the recommended number of poses for a comprehensive conformational search 38 . When analyzing the structure of the protein-peptide complex (Mpro-AGVAKAKAV), we can see that it makes a series of contacts between the different chains, calculated using the COCαDA tool [ 36 , 37 ] (Fig. 5 , Table 2). For example, we can cite the predicted hydrogen bonds between T26 (threonine 26 of the protein) and V9 (valine 9 of the peptide), or a salt bridge between E166 (glutamate 166 of the protein) and K7 (lysine 7 of the peptide). ​​Table 2. Interatomic contacts detected on the interface of the complex Mpro-AGVAKAKAV, calculated using the COCαDA tool [ 36 , 37 ]. # Contact Chain1 R1 Atom1 Chain2 R2 Atom2 Distance 1 T26/V9 A T26 N B V9 O 2.78 A T26 O B V9 N 3.19 2 F140/K7 A F140 O B K7 NZ 2.63 3 G143/K7 A G143 N B K7 O 2.89 4 G143/A8 A G143 N B A8 O 3.68 5 S144/K7 A S144 N B K7 O 3.14 6 C145/K7 A C145 N B K7 O 3.13 A C145 SG B K7 N 3.07 A C145 SG B K7 O 3.55 7 E166/K5 A E166 N B K5 O 2.87 A E166 O B K5 N 2.85 8 E166/K7 A E166 OE1 B K7 NZ 2.83 A E166 OE2 B K7 NZ 2.9 9 Q189/A6 A Q189 OE1 B A6 N 3.29 10 T190/A4 A T190 O B A4 N 2.89 11 Q192/G2 A Q192 N B G2 O 3.73 Also, when we consider all 100 generations, we can see that the protein-peptide binding energy of the models generated by ColabFold was lower for the most part (Fig. 6 ). It is essential to note that the lower the binding energy, the stronger the binding force. However, we can see that, in the first generations, the models proposed by Rosetta had a lower score (for example, generation 1, 4, 5, 6, 8, 11). This suggests that after a certain number of generations, the strategy using ColabFold consistently outperformed Rosetta, as it appeared to model the proposed peptide within the binding site in a more consistent and complementary pose (Fig. 7 ). Regarding the occupancy metric, the best results of each generation varied between 60% and 80%. The occupancy defines which original residues of the Mpro binding site are likely to interact with a residue of the peptide. In the case of the Rosetta experiment, the occupancy values ​​varied throughout the experiment. In the case of the ColabFold experiments, the values ​​increased over the generations. The maximum occupancy observed was 79% (Supplementary Figure S1 ). We also observed that the average size of the peptides with the best docking score decreased over the generations, as shown in the data from the experiment using Rosetta (Fig. 8 , above). For the experiment using ColabFold, the average peptide size was less than 10 amino acids in almost all generations. For over 40 generations, the best peptide had only five amino acids (Fig. 8 , below). Using manual curation to detect a better ligand The results of the case study suggest that the AGVAKAKAV peptide is a potential binder for the Mpro protein. However, analysis of the peptide’s structure indicated that its size could be reduced by removing N-terminal residues, specifically alanine at position 1 and glycine at position 2. The literature has shown that shorter peptides tend to bind better to the receptor [ 39 , 40 ]. Wier & Beekman (2025) suggest that shortening the peptide sequence (truncation), allowing it to contain only residues essential for the interaction, can improve the efficiency of the ligand and may also simplify the synthesis process [ 39 ]. Furthermore, alanine and glycine are small, neutral amino acids that generally contribute little to specific interactions with the protein's active site and may be dispensable for maintaining binding affinity and stability. Therefore, we hypothesized that the VAKAKAV peptide would bind more effectively to Mpro. To assess this, we performed molecular dynamics experiments to verify the binding of the Mpro-N3 complex (the original structure), the Mpro-AGVAKAKAV complex, and the Mpro-VAKAKAV complex. The molecular dynamics results are discussed in the next section. Molecular dynamics simulations To evaluate the structural stability and binding affinity of the complexes formed between SARS-CoV-2 Mpro and the proposed peptides, we performed 100-ns molecular dynamics simulations followed by RMSD, RMSF, hydrogen-bonding interactions, and free-energy calculations using the MM-PBSA method. These parameters enabled us to directly compare the dynamic behavior of the three defined systems (Mpro-AGVAKAKAV, Mpro-VAKAKAV, and the Mpro-N3 control) and relate it to the structural data previously described for the N3 (02JAVLPJE010) inhibitor in wild-type Mpro [ 41 ]. RMSD analysis revealed distinct behaviors between the systems containing different types of ligands (Supplementary Figure S2). The Mpro-N3 system, in which Mpro is associated with a ligand (non-peptide chemical structure), exhibited sharper oscillations and abrupt fluctuations throughout the simulation, suggesting a less stable interaction with the catalytic site. In contrast, the complexes formed with the proposed peptides (Mpro-AGVAKAKAV and Mpro-VAKAKAV) achieved more consistent values ​​after the initial phase, indicating structural stabilization of the protein–ligand complex. Among them, Mpro-VAKAKAV exhibited a lower and more uniform plateau compared to MproAGVAKAKAV, suggesting greater cohesion and better accommodation of the peptide in the active site. RMSF analysis along the Mpro residues showed apparent differences between the systems (Supplementary Figure S3). The Mpro-N3 system, associated with the crystallographic ligand, consistently exhibited higher RMSF values across virtually the entire length of the protein, indicating greater residual mobility and less conformational restriction. In contrast, the complexes formed with the peptides (Mpro-AGVAKAKAV and Mpro-VAKAKAV) presented very similar profiles, with reduced RMSF values ​​across most residues, suggesting that the binding to the peptide promoted local stabilization of the protein. Peaks with greater fluctuation were observed in specific regions, possibly corresponding to flexible loops, but without compromising overall stability. These results indicate that the peptides contribute to stiffening of the active site and adjacent regions, reinforcing the stabilizing role already suggested by the RMSD analyses. Hydrogen bond interactions, as observed over the simulation, revealed differences between the systems (Supplementary Figure S4). Mpro-VAKAKAV shows a more persistent and continuously distributed hydrogen bond throughout the simulation, ensuring greater stability to the complex. Consistent interactions were observed with catalytic residues, including D153, C156, and D155, which remained throughout most of the simulation. Mpro-AGVAKAKAV also established important bonds, such as those involving E166, G143, and H164, but in a slightly less continuous manner compared to Complex 2. The Wild-type system (Mpro + N3) presented a more diffuse pattern, with less regular interactions, mostly displaced to regions peripheral to the catalytic site, such as V157 ↔ K100 and S158 ↔ N151, which do not fully correspond to the hotspots described as critical for Mpro inhibition 10 . These results reinforce that the proposed peptides favor greater stabilization of the active site compared to the N3 wild-ligand, with Mpro-VAKAKAV promoting the most consistent hydrogen bond interactions throughout the simulation. Hydrogen bond pairs throughout the simulation reinforce the differences between the systems (Fig. 9 ). Mpro-AGVAKAKAV complex also exhibited persistent bonds, E166 ↔ K5 and K7 ↔ E166, in addition to interactions with G143 and H164, all of which are important residues of the S1 hotspot, part of the substrate-recognition pocket of the Mpro [ 42 ]. Peptides making interactions with such residues are thus highly likely to inhibit the pocket’s activity [ 41 ]. Mpro-VAKAKAV complex demonstrated consistency, with several residue pairs maintaining high occupancy throughout virtually the entire 100-ns runtime. The interactions with D153, D55, and C156 are notable, remaining active throughout, and demonstrate a continuous and well-distributed contact network within the catalytic site. This behavior suggests that Mpro-VAKAKAV complex promoted a good stabilization of Mpro. The Mpro-N3 system presented a more irregular pattern, with diffusely distributed contacts and less temporal continuity. The main pairs (such as V157 ↔ K100 and S158 ↔ N151) exhibited intermediate occupancy, but without the identical important active site residues observed in the peptides. Thus, the heatmaps corroborate that the proposed peptides interact more stably with the Mpro. The cumulative frequency analysis of hydrogen bonds over the 100 ns simulations revealed notable differences among the three systems evaluated (Supplementary Figure S5). Mpro-AGVAKAKAV complex also exhibited relevant interactions, such as K7 ↔ E166, E166 ↔ K5, and G143 ↔ K7, which correspond to key residues of Mpro. Mpro-VAKAKAV complex presented the highest recurrence values, with emphasis on the interactions K7 ↔ E166, K100 ↔ D155, and D153 ↔ C156, which remained at high frequency throughout practically the entire trajectory. This pattern suggests that the peptide fosters a highly stable contact network, thereby contributing to firm anchoring at the catalytic site. On the other hand, the Wild-type system (Mpro-N3) presented lower frequencies distributed in pairs less central to the catalytic site, notably V157 ↔ K100 and S158 ↔ N151. Although these interactions were consistent, they do not fully reproduce the classic hotspots of protease inhibition, suggesting a less optimized docking pattern. Thus, the frequency data reinforce that the proposed peptides, especially Mpro-VAKAKAV, establish a more robust and persistent hydrogen bond network compared to the reference ligand. Free energy predicted using the MM-PBSA method revealed differences between the systems (Table 3 ). Mpro-VAKAKAV complex presented the most favorable value (− 23.97 kcal/mol), followed by Mpro-AGVAKAKAV complex (− 20.71 kcal/mol), while the Wild system exhibited considerably lower affinity (− 8.88 kcal/mol). Table 3 Binding free energy (ΔGbind) estimated by MM-PBSA for the simulated Mpro systems in complexes with the proposed peptides and with the reference ligand (N3). Free Energy MM/PBSA (Kcal/mol) Mpro-AGVAKAKAV -20,71 Mpro-VAKAKAV -23,97 Mpro-N3 -8,88 The stability observed in the RMSD plots indicated that both peptides conferred greater structural cohesion to Mpro compared to the crystallographic ligand N3, with Mpro-VAKAKAV standing out by reaching more stable plateaus. The RMSF results showed that the peptide complexes significantly reduced residual flexibility, especially in regions close to the catalytic site, suggesting conformational constraint induced by peptide binding. Hydrogen bond analyses complemented these findings: Mpro-VAKAKAV established a more persistent and distributed network of interactions throughout the simulation, including stable contacts with critical residues such as D153, D155, and C156. In contrast, Mpro-AGVAKAKAV complex maintained relevant bonds with E166, G143, and H164, albeit less consistently. The Wild system, in turn, exhibited less regular interactions, with most displaced to regions peripheral to the active site, which was reflected in the less favorable free energy value. Thus, by correlating all analyses, it is clear that the proposed peptides promote greater structural stability, reduce protease flexibility, and establish more robust interaction networks with catalytic residues. Among them, Mpro-VAKAKAV complex stands out as the most promising candidate, as it combines the lowest stable RMSD, reduced local fluctuations, greater hydrogen bond persistence, and the most favorable binding free energy. These results suggest that Mpro-VAKAKAV has greater potential for Mpro inhibition compared to Mpro-AGVAKAKAV and the reference complex. Limitations Despite presenting promising results, this study has some limitations that must be acknowledged. First, the procedure for defining new populations within the genetic algorithm is computationally demanding, particularly when modeling complete protein-peptide complexes using ColabFold. In our experiments, running the genetic algorithm for 100 generations required more than 14 days, even on a server equipped with an Nvidia A100 80GB GPU, 786TB RAM, and an AMD Ryzen Threadripper PRO 5995WX 64-core CPU. On the other hand, the docking-based experiments using the same system were completed in just over one day. Second, the experiments indicated that the AGVAKAKAV peptide exhibited favorable binding to Mpro. However, manual inspection suggested that the two N-terminal residues contributed minimally to protein binding. Based on this observation, we proposed a shorter variant (VAKAKAV), which was subsequently supported by molecular dynamics simulations. This outcome highlights that, while the proposed method can generate promising candidate ligands, expert knowledge remains crucial for refining peptide design. Moreover, the algorithm could be further improved by incorporating the size, physicochemical properties, and chemical nature of amino acids, which could enhance the prioritization of residues contributing most to the interaction energy. Additionally, strategies such as systematically truncating nonessential residues at the peptide termini could be explored to optimize binding affinity and reduce peptide flexibility, ultimately leading to more potent and specific binders Finally, it should be noted that the N3 peptide complexed with Mpro contains non-canonical amino acids. This feature may have influenced the molecular dynamics simulations and binding energy calculations, potentially introducing bias into the evaluation process. Conclusion In this study, we employed a genetic algorithm (GA) approach, integrated with docking-based scoring and AI 3D structure modeling, to design peptide binders targeting the main protease (Mpro). Our results indicate that the approach, called EvoPepFold, was successful in finding peptides with a higher interaction energy between the protein-peptide complex. The optimization process identified promising candidates, including AGVAKAKAV and VAKAKAV, which exhibited improved predicted binding affinities compared to the initial peptide sequence (N3). These results suggest that GA-driven strategies can effectively explore sequence space to identify peptides with favorable interaction patterns at the protein interface. However, some limitations must be acknowledged. The current approach relies exclusively on docking scores (Rosetta energy) as a proxy for binding affinity, which may not fully capture the dynamics and solvation effects present in real biological environments. Future work will address experimental assays to validate the computational predictions and explore alternative score functions to evaluate complexes. We are confident that the insights gained in this study will help drive innovation in peptide engineering, opening new paths for their rational design and therapeutic use. Declarations Competing interests The authors declare no competing interests. Author contributions statement FC developed the scripts and performed the experiments. FC and DM write the manuscript. SCA performed the molecular dynamics simulations. LB, APA, RPL, SCA, LMS, and RCMM edit the manuscript. All authors read and approved the final version of this manuscript. Funding This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) – Finance Code 001. Author Contribution FC developed the scripts and performed the experiments. FC and DM write the manuscript. SCA performed the molecular dynamics simulations. LB, APA, RPL, SCA, LMS, and RCMM edit the manuscript. All authors read and approved the final version of this manuscript. Acknowledgement The authors would like to thank the Brazilian agencies CAPES, CNPQ, and Fapemig. This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) – Finance Code 001. Data Availability Data and scripts are available at [https://github.com/LBS-UFMG/evopepfold](https:/github.com/LBS-UFMG/evopepfold) . References Frappier, V., Duran, M., Keating, A. E. & PixelDB Protein–peptide complexes annotated with structural conservation of the peptide binding mode. Protein Sci. 27 , 276–285 (2018). Angelova, A., Drechsler, M., Garamus, V. M. & Angelov, B. Pep-Lipid Cubosomes and Vesicles Compartmentalized by Micelles from Self-Assembly of Multiple Neuroprotective Building Blocks Including a Large Peptide Hormone PACAP-DHA. ChemNanoMat 5, 1381–1389 (2019). Nissan, N., Allen, M. C., Sabatino, D. & Biggar, K. K. Future Perspective: Harnessing the Power of Artificial Intelligence in the Generation of New Peptide Drugs. Biomolecules 14 , 1303 (2024). Lee, A. C. L., Harris, J. L., Khanna, K. K. & Hong J.-H. A comprehensive review on current advances in peptide drug development and design. Int. J. Mol. Sci. 20 , 2383 (2019). Lau, J. L. & Dunn, M. K. Therapeutic peptides: Historical perspectives, current development trends, and future directions. Bioorg. Med. Chem. 26 , 2700–2707 (2018). Lau, J. L. & Dunn, M. K. Therapeutic peptides: Historical perspectives, current development trends, and future directions. Bioorg. Med. Chem. 26 , 2700–2707 (2018). Hashemi, Z. S. et al. In silico Approaches for the Design and Optimization of Interfering Peptides Against Protein–Protein Interactions. Front Mol. Biosci 8 , (2021). Diakou, I. et al. Novel computational pipelines in antiviral structure–based drug design (Review). Biomed. Rep. 17 , 1–5 (2022). Ibrahim, M. et al. Why is the Omicron main protease of SARS-CoV-2 less stable than its wild-type counterpart? A crystallographic, biophysical, and theoretical study. hLife 2 , 419–433 (2024). Sacco, M. D. et al. Structure and inhibition of the SARS-CoV-2 main protease reveal strategy for developing dual inhibitors against Mpro and cathepsin L. Sci. Adv. 6 , eabe0751 (2020). Nissan, N., Allen, M. C., Sabatino, D. & Biggar, K. K. Future Perspective: Harnessing the Power of Artificial Intelligence in the Generation of New Peptide Drugs. Biomolecules 14 , 1303 (2024). Yan, Y., Zhang, D., Zhou, P., Li, B. & Huang, S. Y. HDOCK: a web server for protein–protein and protein–DNA/RNA docking based on a hybrid strategy. Nucleic Acids Res. 45 , W365–W373 (2017). Yan, Y., Tao, H., He, J. & Huang, S. Y. The HDOCK server for integrated protein–protein docking. Nat. Protoc. 15 , 1829–1852 (2020). Zhou, P., Jin, B., Li, H. & Huang, S. Y. HPEPDOCK: a web server for blind peptide–protein docking based on a hierarchical algorithm. Nucleic Acids Res. 46 , W443–W450 (2018). Raveh, B., London, N., Zimmerman, L. & Schueler-Furman, O. Rosetta FlexPepDock ab-initio: simultaneous folding, docking and refinement of peptides onto their receptors. PloS One . 6 , e18934 (2011). Chaudhury, S. et al. Benchmarking and analysis of protein docking performance in Rosetta v3. 2. PloS One . 6 , e22477 (2011). Evans, R. et al. Protein complex prediction with AlphaFold-Multimer. biorxiv 2021–10 (2021). Omidi, A., Møller, M. H., Malhis, N., Bui, J. M. & Gsponer, J. AlphaFold-Multimer accurately captures interactions and dynamics of intrinsically disordered protein regions. Proc. Natl. Acad. Sci. 121, e2406407121 (2024). Fjell, C. D., Jenssen, H., Cheung, W. A., Hancock, R. E. W. & Cherkasov, A. Optimization of antibacterial peptides by genetic algorithms and cheminformatics. Chem. Biol. Drug Des. 77 , 48–56 (2011). Borkakoti, N. & Thornton, J. M. AlphaFold2 protein structure prediction: Implications for drug discovery. Curr. Opin. Struct. Biol. 78 , 102526 (2023). Raveh, B., London, N., Zimmerman, L. & Schueler-Furman, O. Rosetta FlexPepDock ab-initio: Simultaneous Folding, Docking and Refinement of Peptides onto Their Receptors. PLOS ONE . 6 , e18934 (2011). Chaudhury, S. et al. Benchmarking and Analysis of Protein Docking Performance in Rosetta v3.2. PLOS ONE . 6 , e22477 (2011). Raveh, B., London, N., Zimmerman, L. & Schueler-Furman, O. Rosetta FlexPepDock ab-initio: Simultaneous Folding, Docking and Refinement of Peptides onto Their Receptors. PLOS ONE . 6 , e18934 (2011). Berman, H. M. et al. The Protein Data Bank. Nucleic Acids Res. 28 , 235–242 (2000). Martins, P. M. et al. Propedia: a database for protein–peptide identification based on a hybrid clustering algorithm. BMC Bioinform. 22 , 1 (2021). Martins, P. et al. Propedia v2.3: A novel representation approach for the peptide-protein interaction database using graph-based structural signatures. Front Bioinforma 3 , (2023). Mirdita, M. et al. ColabFold: making protein folding accessible to all. Nat. Methods . 19 , 679–682 (2022). Adolf-Bryfogle, J. & Jr, R. L. D. The PyRosetta Toolkit: A Graphical User Interface for the Rosetta Software Suite. PLOS ONE . 8 , e66856 (2013). Abraham, M. J. et al. High performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX 1 . GROMACS , 19–25 (2015). Huang, J. et al. CHARMM36m: an improved force field for folded and intrinsically disordered proteins. Nat. Methods . 14 , 71–73 (2017). Malde, A. K. et al. An automated force field topology builder (ATB) and repository: version 1.0. J. Chem. Theory Comput. 7 , 4026–4037 (2011). Michaud-Agrawal, N., Denning, E. J., Woolf, T. B. & Beckstein, O. MDAnalysis: a toolkit for the analysis of molecular dynamics simulations. J. Comput. Chem. 32 , 2319–2327 (2011). Gowers, R. J. et al. MDAnalysis: a Python package for the rapid analysis of molecular dynamics simulations. (Los Alamos National Laboratory (LANL), Los Alamos, NM (United States), (2019). Hunter, J. D. & Matplotlib A 2D graphics environment. Comput. Sci. Eng. 9 , 90–95 (2007). Valdés-Tresanco, M. S., Valdés-Tresanco, M. E., Valiente, P. A. & Moreno, E. gmx_MMPBSA: a new tool to perform end-state free energy calculations with GROMACS. J. Chem. Theory Comput. 17 , 6281–6291 (2021). Lemos, R. P., Mariano, D., de Silveira, S., de Melo-Minardi, R. C. & A. & COCαDA-A Fast and Scalable Algorithm for Interatomic Contact Detection in Proteins Using Cα Distance Matrices. Front. Bioinforma . 5 , 1630078 (2025). Lemos, R., Mariano, D., Silveira, S. & Melo-Minardi, R. COCαDA - Large-Scale Protein Interatomic Contact Cutoff Optimization by Cα Distance Matrices. 70 (2024). 10.5753/bsb.2024.245545 Meng, E. C. et al. UCSF ChimeraX: Tools for structure building and analysis. Protein Sci. 32 , e4792 (2023). van Wier, S. P. & Beekman, M. A. Peptide design to control protein–protein interactions. (2025). 10.1039/D4CS00243A Leenheer, D., ten Dijke, P. & Hipolito, C. J. A current perspective on applications of macrocyclic-peptide-based high-affinity ligands. Pept. Sci. 106 , 889–900 (2016). Jin, Z. et al. Structure of Mpro from SARS-CoV-2 and discovery of its inhibitors. Nature 582 , 289–293 (2020). Verma, N., Henderson, J. A. & Shen, J. Proton-Coupled Conformational Activation of SARS Coronavirus Main Proteases and Opportunity for Designing Small-Molecule Broad-Spectrum Targeted Covalent Inhibitors. J. Am. Chem. Soc. 142 , 21883–21890 (2020). Additional Declarations No competing interests reported. Supplementary Files supplementarymaterial1.pdf Cite Share Download PDF Status: Published Journal Publication published 29 Nov, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 08 Oct, 2025 Reviews received at journal 06 Oct, 2025 Reviews received at journal 06 Oct, 2025 Reviews received at journal 03 Oct, 2025 Reviewers agreed at journal 02 Oct, 2025 Reviewers agreed at journal 02 Oct, 2025 Reviewers agreed at journal 01 Oct, 2025 Reviewers agreed at journal 30 Sep, 2025 Reviewers agreed at journal 30 Sep, 2025 Reviewers invited by journal 29 Sep, 2025 Editor invited by journal 29 Sep, 2025 Editor assigned by journal 26 Sep, 2025 Submission checks completed at journal 26 Sep, 2025 First submitted to journal 24 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7706745","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":526299469,"identity":"84becb40-5a51-4ddf-9546-b60d206dc712","order_by":0,"name":"Frederico Chaves","email":"","orcid":"","institution":"Universidade Federal de Minas Gerais","correspondingAuthor":false,"prefix":"","firstName":"Frederico","middleName":"","lastName":"Chaves","suffix":""},{"id":526299470,"identity":"464c3323-f2d3-437a-a2eb-dc8d7395eac1","order_by":1,"name":"Diego Mariano","email":"","orcid":"","institution":"Universidade Federal de Minas Gerais","correspondingAuthor":false,"prefix":"","firstName":"Diego","middleName":"","lastName":"Mariano","suffix":""},{"id":526299472,"identity":"5abd9b0c-8aee-4f16-9573-3dee7fb8a377","order_by":2,"name":"Luana Bastos","email":"","orcid":"","institution":"Universidade Federal de Minas Gerais","correspondingAuthor":false,"prefix":"","firstName":"Luana","middleName":"","lastName":"Bastos","suffix":""},{"id":526299473,"identity":"c2b1bb6b-c83e-4711-b385-bf077ef8a437","order_by":3,"name":"Ana Paula Abreu","email":"","orcid":"","institution":"Universidade Federal de Minas Gerais","correspondingAuthor":false,"prefix":"","firstName":"Ana","middleName":"Paula","lastName":"Abreu","suffix":""},{"id":526299475,"identity":"cf6f8326-1840-496b-96d9-0560b931158e","order_by":4,"name":"Rafael Pereira Lemos","email":"","orcid":"","institution":"Universidade Federal de Minas Gerais","correspondingAuthor":false,"prefix":"","firstName":"Rafael","middleName":"Pereira","lastName":"Lemos","suffix":""},{"id":526299479,"identity":"2f266e8f-3393-4bbb-ad44-55a1a6365108","order_by":5,"name":"Sheila Cruz Araújo","email":"","orcid":"","institution":"Universidade Federal de Minas Gerais","correspondingAuthor":false,"prefix":"","firstName":"Sheila","middleName":"Cruz","lastName":"Araújo","suffix":""},{"id":526299481,"identity":"67c016da-f34e-4989-8604-0eb0261ae2fd","order_by":6,"name":"Lucas Moraes Santos","email":"","orcid":"","institution":"Universidade Federal de Minas Gerais","correspondingAuthor":false,"prefix":"","firstName":"Lucas","middleName":"Moraes","lastName":"Santos","suffix":""},{"id":526299482,"identity":"2dd09342-b323-446e-a259-80dee74a71b4","order_by":7,"name":"Raquel Cardoso Melo-Minardi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA20lEQVRIiWNgGAWjYNCCAgYGfoYEEEsOiHmI0WLAwCDZANZiTIIWgwPEajFvP7xN4oOBXbTx8eSHHxjbDKLlHXiPfcCnReZMWpnkDIPk3G1nnhlLALXkbjzAlzwDnxYJhhwzaR6DA7nbbuSwMTC2/cnd2MBjjNdhEvxvzKT/ALVsngHWYkCEFgmgLUC/526QgGqZz0BQy7Niyx6gX2aA/JJwziB3AzNfMgGHJW+88aPCLre/HRhiH8qAtrT3HsarhQEcKTCQAOIS1ICiBQTkGwhrGQWjYBSMgpEFAIltRKjJZpMXAAAAAElFTkSuQmCC","orcid":"","institution":"Universidade Federal de Minas Gerais","correspondingAuthor":true,"prefix":"","firstName":"Raquel","middleName":"Cardoso","lastName":"Melo-Minardi","suffix":""}],"badges":[],"createdAt":"2025-09-24 20:08:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7706745/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7706745/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-28061-y","type":"published","date":"2025-11-29T15:58:37+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":93620907,"identity":"5a6cd395-2a13-445a-a1fa-6b0186e51b0d","added_by":"auto","created_at":"2025-10-15 17:57:43","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10567915,"visible":true,"origin":"","legend":"","description":"","filename":"artigofredevopepfold3.docx","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/7bbca082fadb7a6d7a969424.docx"},{"id":93621626,"identity":"42de37bd-d31a-44df-8b6c-f7df361608b9","added_by":"auto","created_at":"2025-10-15 18:05:43","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8720,"visible":true,"origin":"","legend":"","description":"","filename":"1b504d95fd9f492b86e33abfebbd57f9.json","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/08a1211ab4947aeed4c4c363.json"},{"id":93621641,"identity":"11349fc4-e042-4642-b6c4-d7044b7ea9e6","added_by":"auto","created_at":"2025-10-15 18:05:45","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1025423,"visible":true,"origin":"","legend":"","description":"","filename":"supplementarymaterial1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/4def7cd9cab9d5bc24552e32.pdf"},{"id":93621625,"identity":"f0aa29fc-2227-4e29-ad45-9d626fdc950e","added_by":"auto","created_at":"2025-10-15 18:05:43","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":129568,"visible":true,"origin":"","legend":"","description":"","filename":"1b504d95fd9f492b86e33abfebbd57f91enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/cfb04cb13d2ba63c23aae92d.xml"},{"id":93620884,"identity":"58850eb6-c09b-4c97-acd4-e8c6ee01c477","added_by":"auto","created_at":"2025-10-15 17:57:43","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2778811,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/3ec8d398ada0b450e74ee7ad.png"},{"id":93620887,"identity":"8219b2d9-a0ce-413c-bc67-746ab7a03056","added_by":"auto","created_at":"2025-10-15 17:57:43","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":512577,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/20975f914aea814772b5618f.png"},{"id":93621628,"identity":"a360234b-ef37-4f64-b0d5-b928f9ef6122","added_by":"auto","created_at":"2025-10-15 18:05:43","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":37735,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/7e14bc595e8a038b36e01fed.png"},{"id":93621640,"identity":"e76b5789-13fd-4226-8c78-0d463a9c81c6","added_by":"auto","created_at":"2025-10-15 18:05:45","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":454439,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/e3609144b24a67212327ca27.jpeg"},{"id":93621629,"identity":"4e2fcd00-d23a-4325-ab23-568aa067446b","added_by":"auto","created_at":"2025-10-15 18:05:43","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1853253,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/78043d9f670bf3a4b6f7b7ed.png"},{"id":93621632,"identity":"827f396a-6a4f-4aac-ba5a-bed29727b819","added_by":"auto","created_at":"2025-10-15 18:05:43","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":47743,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/5cf24b20ca5229742bb31be8.png"},{"id":93620888,"identity":"12f573fb-9405-4293-ad42-96d85ff6565a","added_by":"auto","created_at":"2025-10-15 17:57:43","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":94592,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/6c922f2b18da9b33e946d248.png"},{"id":93621636,"identity":"f529eabd-fe4a-4a6e-8ca5-bca0dcb04956","added_by":"auto","created_at":"2025-10-15 18:05:43","extension":"jpeg","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":435020,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage8.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/a4a7ad1e91ebcf83d1e828f1.jpeg"},{"id":93620905,"identity":"105e44ac-199f-4f1d-9746-c5a986e52a7c","added_by":"auto","created_at":"2025-10-15 17:57:43","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":544522,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/f7536e3f9ea4247a75672bc2.png"},{"id":93620909,"identity":"ab443922-56e9-449a-8eef-2a1062f17008","added_by":"auto","created_at":"2025-10-15 17:57:43","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":353145,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/e1bc09f6b2221555703ebe1b.png"},{"id":93622881,"identity":"f8a38675-2c36-4d73-9e04-d493126b2c1a","added_by":"auto","created_at":"2025-10-15 18:21:43","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":116299,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/9766bcc3d5444cdebd3c9184.png"},{"id":93620897,"identity":"ec7b731a-d09a-459b-b969-f53f5227d5c0","added_by":"auto","created_at":"2025-10-15 17:57:43","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18395,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/ff9af0542a5decd1ed327c43.png"},{"id":93620902,"identity":"57a78503-9b9d-421d-ace3-eb88c1f55cd7","added_by":"auto","created_at":"2025-10-15 17:57:43","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":436327,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/57b020df1c9907d5c2c000a8.png"},{"id":93621637,"identity":"18ce58f1-b2ec-4333-abff-8f931937cecf","added_by":"auto","created_at":"2025-10-15 18:05:43","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":233340,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/04346a3a76f4326658a5e7fb.png"},{"id":93621634,"identity":"92285cdf-eb37-4b26-9ca6-be59033e9be1","added_by":"auto","created_at":"2025-10-15 18:05:43","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19458,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/ee449432b2ad009e37fe4e70.png"},{"id":93620906,"identity":"e75439ca-2159-4780-b430-fa865381d973","added_by":"auto","created_at":"2025-10-15 17:57:43","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":28992,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/0e5a0ad59a7d4653f12dc134.png"},{"id":93622572,"identity":"c004e700-6cf0-4626-9ba2-f47c2d1da54c","added_by":"auto","created_at":"2025-10-15 18:13:43","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":73148,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/7800d3534c99f3f359520160.png"},{"id":93620895,"identity":"20fe47d7-1ee9-431b-b285-51c3fdd964b5","added_by":"auto","created_at":"2025-10-15 17:57:43","extension":"png","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":118068,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/73d6d0b38ce54cd82a1d21b8.png"},{"id":93620910,"identity":"9b0e0fa2-4dd1-44da-9474-982749b621dc","added_by":"auto","created_at":"2025-10-15 17:57:43","extension":"xml","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":128597,"visible":true,"origin":"","legend":"","description":"","filename":"1b504d95fd9f492b86e33abfebbd57f91structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/fec76a9294aaf82e1b468a31.xml"},{"id":93620908,"identity":"46095baa-5c48-489e-af40-bf215a59a0e6","added_by":"auto","created_at":"2025-10-15 17:57:43","extension":"html","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":140683,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/b72624d27cb86fcae0ead6f9.html"},{"id":93620880,"identity":"b33727df-2040-49cb-b802-cc09d921e8cb","added_by":"auto","created_at":"2025-10-15 17:57:43","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1118739,"visible":true,"origin":"","legend":"\u003cp\u003eTarget structure. (A) COVID-19 main protease (Mpro – purple surface) complexed with the inhibitor N3 (green) (PDB ID: 6LU7). (B) MPRO residues directly interacting with the N3 peptide. (C) We selected residues up to 5Å from the N3 peptide (in orange). (D) Then, we removed the peptide from the complex. (E) MPRO docking target site residues: T24, T25, T26, L27, H41, M49, F140, L141, N142, G143, S144, C145, H164, M165, E166, L167, P168, H172, D187, R188, Q189, T190, A191, and Q192.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/2e5c21972ac784535b62f9f0.png"},{"id":93620876,"identity":"0f07cc3c-b0fd-4042-b4f4-58ab68ca7386","added_by":"auto","created_at":"2025-10-15 17:57:43","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":817207,"visible":true,"origin":"","legend":"\u003cp\u003eOverview of the genetic algorithm. Peptides obtained from the Propedia database were initially docked to the Mpro structure. The top 100 results were selected as the initial population for the genetic algorithm. Two strategies were then employed: a docking-based approach (utilizing the Rosetta suite) and an AI-driven 3D modeling approach (utilizing ColabFold). A fitness function evaluated peptide candidates through a tournament selection scheme, and genetic operations were applied to peptide sequences to generate a new population. Each new set of peptide structures was modeled through docking (Rosetta) or 3D modeling (ColabFold), resulting in a population of 25 structures. These steps were iterated over 100 generations. Finally, the best peptides from each generation were selected based on the lowest docking scores.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/8dfdec419e424a4678166fd6.png"},{"id":93621622,"identity":"d7bd959e-75ab-411c-8f19-082622cd2222","added_by":"auto","created_at":"2025-10-15 18:05:43","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":259240,"visible":true,"origin":"","legend":"\u003cp\u003eExamples of mutations of the sequence KWGTSHVF: insertion, deletion, and substitution. Additionally, a crossing over between this sequence and QYADREMP is shown on the right.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/ebd41d164caa645ef16e4c90.png"},{"id":93622567,"identity":"00755ae3-0116-48dc-98b8-c404660993c6","added_by":"auto","created_at":"2025-10-15 18:13:43","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":454439,"visible":true,"origin":"","legend":"\u003cp\u003eBest protein-peptide complexes for generations: G1, G10, G25, G50, G75, G84, G96, and G100. The star indicates the generations that showed the best global complexes in the Rosetta (G96) and ColabFold (G84) experiments. Scores are given as REU (Rosetta Energy Unit) values.\u003c/p\u003e","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/b6d24a04566993b9f001bf9a.jpeg"},{"id":93620882,"identity":"4187ace2-a1cf-4a7e-9c86-ef65b6466ec4","added_by":"auto","created_at":"2025-10-15 17:57:43","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1048787,"visible":true,"origin":"","legend":"\u003cp\u003ePeptide AGVAKAKAV (sticks with surface displayed in green) complexed with Mpro (purple cartoon). Contacts are displayed with dashed blue lines. Figure generated using ChimeraX [38].\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/eee2ade19911748c2301888e.png"},{"id":93622568,"identity":"18ea50fc-8075-48c6-b7e0-47287549fac2","added_by":"auto","created_at":"2025-10-15 18:13:43","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":123750,"visible":true,"origin":"","legend":"\u003cp\u003eLowest value of docking energy score for each generation of Rosetta (blue line) and ColabFold (green line) experiments.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/9ad0d44009e4f0f3c7c97c96.png"},{"id":93622569,"identity":"64b34e53-07b8-4c59-88f0-bead65ccc122","added_by":"auto","created_at":"2025-10-15 18:13:43","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":341811,"visible":true,"origin":"","legend":"\u003cp\u003eDifference in docking scores between Rosetta and ColabFold for the best results of each of the 100 generations. Red bars indicate that Rosetta's approach obtained better performance. Blue bars indicate ColabFold approach obtained a better performance.\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/2c540664f050f5fedfc73914.png"},{"id":93622571,"identity":"09cb164e-3479-478a-83d4-de63c2524488","added_by":"auto","created_at":"2025-10-15 18:13:43","extension":"jpeg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":435020,"visible":true,"origin":"","legend":"\u003cp\u003eLength distribution of the best peptide of each generation (total amino acids).\u003c/p\u003e","description":"","filename":"floatimage8.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/0f0c7a115c1d4b3e10d4bafc.jpeg"},{"id":93620894,"identity":"c3be43b5-7588-4a13-812f-3760433190a6","added_by":"auto","created_at":"2025-10-15 17:57:43","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":1386929,"visible":true,"origin":"","legend":"\u003cp\u003eHeatmaps of the ten most frequent hydrogen bonding interactions over 100 ns of simulation. Color intensity reflects bond occupancy at each time interval (dark blue = 100% presence; light yellow = absence).\u003c/p\u003e","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/5f6aa4de7143354c8636e509.png"},{"id":97179475,"identity":"4d596cbc-9a94-4333-947d-321b27e61bb2","added_by":"auto","created_at":"2025-12-01 16:15:52","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":6924591,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/36359345-2525-44d7-938a-73cf708986bc.pdf"},{"id":93621627,"identity":"4bc9ee2b-c435-40cd-abd0-1935de403e61","added_by":"auto","created_at":"2025-10-15 18:05:43","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1025423,"visible":true,"origin":"","legend":"","description":"","filename":"supplementarymaterial1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7706745/v1/e6ba0ed84fb9ea61fe74b3ea.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"EvoPepFold: A Hybrid Evolutionary and Structural Pipeline for AI- Guided Peptide Inhibitor Design Using AlphaFold and Rosetta","fulltext":[{"header":"Introduction","content":"\u003cp\u003ePeptides are molecules composed of short chains of amino acids linked by peptide bonds, typically consisting of 2\u0026ndash;50 residues [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. In living beings, they play diverse roles in biological systems, acting as hormones, signaling molecules, neurotransmitters, and regulators of various physiological processes [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. For example, in the case of infectious diseases, peptides can function as antiviral agents by interfering with key stages of the viral life cycle, such as entry, replication, or assembly. Additionally, peptide inhibitors represent a promising class of antiviral therapeutics, offering several advantages over traditional small-molecule drugs [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eAdvances in synthetic approaches have enabled the modification of the biophysical and biochemical properties of peptides, making them promising candidates for drug development, mainly due to their low toxicity, high specificity, and biocompatibility [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Currently, more than 60 peptide-based drugs have been approved, with many others in clinical trials [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. These compounds have demonstrated effectiveness in treating diseases such as cancer, type 2 diabetes, and autoimmune disorders, with exenatide derivatives standing out as notable examples [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Furthermore, understanding protein\u0026ndash;peptide interactions is essential for the rational design of new compounds with therapeutic and biotechnological potential [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eAntiviral peptides can be strategically developed to bind to vital viral enzymes or structural components, thereby blocking their activity. As a result, there is growing interest in the pharmaceutical industry in creating peptides capable of interfering with essential protein\u0026ndash;protein interactions required for viral replication [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. These peptides emerged as promising allies in the fight against viral infections, such as COVID-19.\u003c/p\u003e\u003cp\u003eThreats to public health, such as the SARS-CoV-2 virus, highlight the need for methodologies to expedite the development of effective antiviral therapeutics [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. While vaccination efforts have significantly mitigated the severity of the pandemic, the emergence of new variants and the need for treatments for infected individuals underscore the importance of antiviral drug discovery. In the particular case of SARS-CoV-2, the main protease (Mpro) is a crucial enzyme responsible for processing viral polyproteins. It has emerged as a prime target for therapeutic intervention due to its essential role in viral replication and its distinct substrate specificity compared to human proteases [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. The unique preference of Mpro for a glutamine residue at the P1 position of its substrates presents an opportunity to design highly selective inhibitors [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. However, the potential for drug resistance arising from mutations in the Mpro sequence highlights the need for innovative approaches to identify novel antiviral agents [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eAs discussed, peptide inhibitors represent a promising class of antiviral therapeutics, offering several advantages over traditional small-molecule drugs [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Their inherent ability to interact with large protein surfaces with high specificity and potency makes them well-suited for targeting enzymes like Mpro [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. However, designing peptides with high affinity to specific protein complexes is not trivial.\u003c/p\u003e\u003cp\u003eIn this context, computational strategies have been widely employed to aid in designing peptides with enhanced affinity for the receptor. For example, molecular docking methods, such as HDOCK [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e], HPEPDOCK [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], and Rosetta [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e], can help in identifying the binding poses of the peptide to the protein and predicting their binding energy. Although not explicitly designed for this purpose, AI-based structural modeling tools like AlphaFold Multimer [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] have also been adopted to simulate protein-peptide interactions [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. However, these tools do not perform regular molecular docking; instead, they model the entire complex using deep neural network techniques trained on large datasets of protein-protein interactions.\u003c/p\u003e\u003cp\u003eFurthermore, various computational methods can be employed to automate and optimize the generation of peptides with therapeutic properties. Genetic algorithms (GAs) consist of one class of such optimization methods. They were developed inspired by evolution, which iteratively evolve a population of candidate solutions to improve their performance according to a defined fitness function. The use of these algorithms for sequence-based peptide optimization is already widely discussed in the literature [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]; however, to the best of our knowledge, the combination of molecular docking, AI-based modeling, molecular dynamics simulations, and genetic algorithms to optimize peptides has not yet been explored.\u003c/p\u003e\u003cp\u003eIn this study, we aimed to identify peptides with the potential to bind to Mpro, thereby helping to develop new therapeutic candidates against SARS-CoV-2. To achieve this, we developed and evaluated the feasibility of an innovative computational pipeline composed of three main steps: (i) a genetic algorithm for the generation and optimization of peptide sequences based on evolutionary criteria; (ii) AlphaFold2 [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e], a deep learning-based tool used to predict the three-dimensional structures of the generated peptides; and (iii) the Rosetta software [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] using its scoring function [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] to estimate the binding energy between each peptide and Mpro. The function estimates this energy based on physicochemical criteria, serving as a relative indicator of complex stability and aiding in the selection of the most promising peptides. Lastly, we evaluated the best proposed peptides using molecular dynamics simulations.\u003c/p\u003e"},{"header":"Material and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eData collection\u003c/h2\u003e\u003cp\u003eWe collected, from the Protein Data Bank (PDB) [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e], the 3D structure of the COVID-19 main protease (Mpro) in complex with an inhibitor N3 peptide (PDB ID: 6LU7) and defined this structure as the target. N3 is a peptide of six amino acid residues (sequence: \u0026ldquo;XAVLXX\u0026rdquo;, where X corresponds to a non-canonical amino acid).\u003c/p\u003e\u003cp\u003eAdditionally, we collected 2,355 peptides composed of 5 to 30 amino acids from the Propedia database [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. These peptides were used to define the initial population for the genetic algorithm to improve on.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eMpro docking site definition\u003c/h3\u003e\n\u003cp\u003eWe used the binding position of the N3 peptide in the Mpro to define the binding region of interest (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA-B). We selected residues with at least one atom within 5 \u0026Aring; of the N3 peptide, and described this region as the docking site contact interface (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC). Finally, we removed the peptide from the complex (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD) and used the remaining structure as a target for further analyses. The following 24 residues were selected from the Mpro interface of contact with the N3 peptide (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE): T24, T25, T26, L27, H41, M49, F140, L141, N142, G143, S144, C145, H164, M165, E166, L167, P168, H172, D187, R188, Q189, T190, A191, and Q192.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eOur ultimate goal is to propose a peptide that binds to the Mpro structure with more affinity than the N3 peptide. Therefore, this hypothetical peptide should at least occupy the same binding site as the original peptide. To measure this, we define the residue occupancy (RO) parameter. The RO parameter indicates the percentage of the 24 amino acids of Mpro at least 5\u0026Aring; away from the ligand. By default, the original peptide has RO\u0026thinsp;=\u0026thinsp;100%. A docked peptide with an RO score equal to 0% is bound to a binding site different from the one targeted in this work.\u003c/p\u003e\n\u003ch3\u003eGenetic algorithm overview\u003c/h3\u003e\n\u003cp\u003e\u003c/p\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003e. Overview of the genetic algorithm. Peptides obtained from the Propedia database were initially docked to the Mpro structure. The top 100 results were selected as the initial population for the genetic algorithm. Two strategies were then employed: a docking-based approach (utilizing the Rosetta suite) and an AI-driven 3D modeling approach (utilizing ColabFold). A fitness function evaluated peptide candidates through a tournament selection scheme, and genetic operations were applied to peptide sequences to generate a new population. Each new set of peptide structures was modeled through docking (Rosetta) or 3D modeling (ColabFold), resulting in a population of 25 structures. These steps were iterated over 100 generations. Finally, the best peptides from each generation were selected based on the lowest docking scores.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe evolutionary process began with an initial population of 100 peptides. In each generation, new peptides were produced through two main genetic operators: \u003cem\u003ecrossover\u003c/em\u003e and \u003cem\u003emutation\u003c/em\u003e. The crossover operator recombined segments of parent peptides using variable crossover lengths, which were randomly chosen for each operation. The mutation operator introduced diversity by applying insertion, deletion, or substitution of amino acid residues at random positions in the peptide sequence.\u003c/p\u003e\u003cp\u003eNewly generated peptides were modeled using two strategies: docking protein-peptide using Rosetta [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] and AI-modelling using ColabFold, an AlphaFold2-powered tool [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] (further details are provided in the following subsections).\u003c/p\u003e\u003cp\u003eThe \u003cem\u003efitness\u003c/em\u003e function used to evaluate each peptide was the docking score; however, the occupancy of the binding site was used to filter out peptides that bind elsewhere. The highest-ranking peptides were selected to propagate the next generation. This iterative evolutionary process was repeated until convergence criteria were met, aiming to identify peptides with enhanced binding potential to Mpro.\u003c/p\u003e\n\u003ch3\u003eInitial population\u003c/h3\u003e\n\u003cp\u003eTo define the initial population, docking was performed between the 2,355 peptides collected from Propedia and the Mpro structure using PyRosetta4 - release 2024.39 [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Due to the large number of peptides to be tested, we performed low-resolution \u003cem\u003eab initio\u003c/em\u003e docking, generating five poses for each peptide, followed by an energy minimization step. To accelerate computation, docking jobs were run in parallel on 50 CPUs (50 peptides processed concurrently, one peptide per CPU). Peptides that did not achieve at least 30% RO were removed from further consideration. The remaining peptides were ranked by their best docking score across the five poses, and the top candidates were selected for downstream analysis.\u003c/p\u003e\n\u003ch3\u003eParameters definition\u003c/h3\u003e\n\u003cp\u003eTo define appropriate genetic algorithm parameters, a preliminary tuning run was performed using the top 25 peptides from the initial population. For each tested parameter combination, the GA was executed for 20 generations using a reduced-fidelity setup, where only one docking pose per peptide was computed using Rosetta [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Although this approach does not adhere to best practices for docking-based scoring, it significantly reduces computational cost and enables the rapid comparison of parameter effectiveness. The goal of this experiment was not to estimate binding affinity, but to evaluate the relative impact of key parameters, including mutation rate, tournament size, crossover rate, and elitism, on optimization performance. The best-performing parameter set from this coarse search was then adopted for the full-scale pipeline, where higher-resolution scoring was applied.\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eOperations\u003c/h2\u003e\u003cp\u003eThe population was subjected to two types of genetic operations: crossover and mutation (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e3\u003c/span\u003e). In the crossover operation, peptide segments were exchanged between two parent sequences, generating new peptide variants. In the mutation operation, diversity was introduced through random substitution, deletion, or insertion of amino acid residues. The type of operation applied at each step was chosen probabilistically, with 90% chance of performing crossover and a 10% chance of performing mutation.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eStructural modeling\u003c/h3\u003e\n\u003cp\u003eIn the next phase, two modeling strategies were employed to evaluate peptide\u0026ndash;Mpro binding. In the first approach, each sequence was modeled as a random conformation peptide and docked to the Mpro using Rosetta\u0026rsquo;s FlexPepDock [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] high-resolution \u003cem\u003eab initio\u003c/em\u003e protocol, generating ten poses per peptide. In the second approach, AlphaFold2 [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] (in multimer mode) was used to directly predict the structure of the peptide\u0026ndash;Mpro complex.\u003c/p\u003e\u003cp\u003eBoth experiments were executed on the same server \u0026mdash; an AMD Ryzen Threadripper PRO 5995WX 64-Cores processor equipped with an 80GB NVIDIA A100 GPU. For each evaluation, the total time per generation and the time per peptide were recorded. In the AlphaFold2 run, multiple sequence alignments were generated using ColabFold [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e], and structure prediction was performed locally. Due to resource constraints, a 90-minute break was added between generations to avoid GPU saturation. In the Rosetta run, 25 CPU cores were used to perform docking in parallel, allowing batches of 25 peptides to be processed concurrently.\u003c/p\u003e\n\u003ch3\u003eFitness function\u003c/h3\u003e\n\u003cp\u003eThe top-performing peptides were identified using a tournament-based selection strategy. Complexes generated from both approaches were evaluated with Rosetta\u0026rsquo;s energy function [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e], which estimates binding free energy at the peptide\u0026ndash;protein interface. The PyRosetta library [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e] was used to assess binding energy and automate the score calculations. These scores guided the GA. In each variation, the best-performing peptide in each generation was retained via elitism. At the same time, the remaining population underwent tournament selection, mutation, and crossover to create the next generation of sequences. The best peptide can be maintained for up to three generations. This process was repeated for 100 generations, allowing the algorithm to converge on peptides with progressively improved predicted binding characteristics.\u003c/p\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eMolecular dynamics simulations\u003c/h2\u003e\u003cp\u003eTo evaluate the best results proposed by the case study, we performed molecular dynamics (MD) simulation experiments. The simulations were performed using GROMACS [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e] with the CHARMM36 force field [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e] and a standard explicit water model on a workstation equipped with CUDA-enabled GPU acceleration (Nvidia A100 80GB). Protein\u0026ndash;ligand complexes were placed in a cubic box, centered, solvated, and neutralized with counterions. Energy minimization was performed using the steepest-descent algorithm (50,000 steps). Equilibration proceeded in two stages: NVT for 100 ps at 300 K with a V-rescale thermostat, followed by NPT for 100 ps at 300 K and 1 bar with a V-rescale thermostat and a Parrinello\u0026ndash;Rahman barostat, with protein atoms restrained. The production run lasted 100 ns (50,000,000 steps; 2 fs timestep) at 300 K and 1 bar, using the Verlet cutoff scheme and hydrogen-bond constraints.\u003c/p\u003e\u003cp\u003eTrajectories were centered and then least-squares-fitted on protein backbone atoms prior to analysis. RMSD (Root Mean Square Deviation) and RMSF (Root Mean Square Fluctuation) were computed with GROMACS Tools and the MDAnalysis Python library [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]; plots were generated with Matplotlib [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eBinding free energies (ΔG_bind) were estimated by MM-PBSA using gmx_MMPBSA [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Polar solvation energies were obtained by solving the Poisson\u0026ndash;Boltzmann equation, and non-polar contributions were estimated from solvent-accessible surface area (SASA). Energies for complex, receptor, and ligand were evaluated over uniformly spaced frames extracted from the production trajectory, and ΔG_bind was computed as ΔG_complex\u0026thinsp;\u0026minus;\u0026thinsp;ΔG_receptor\u0026thinsp;\u0026minus;\u0026thinsp;ΔG_ligand, with ensemble averaging. Per-residue free-energy decomposition was also performed.\u003c/p\u003e\u003c/div\u003e"},{"header":"Results and discussion","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003ePeptides proposed to Mpro\u003c/h2\u003e\u003cp\u003eIn the performed experiment, 50 peptides were proposed for each generation of the genetic algorithm. The best peptide is defined based on two metrics: docking energy score of Rosetta (the lower, the better) and occupancy (the higher, the better). Rosetta energies are given on a scale named Rosetta Energy Unit (REU), which is derived from a combination of physics-based and statistical potentials. Figure\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e4\u003c/span\u003e shows the complexes formed by the binding of Mpro to the peptide that obtained the lowest docking score, and the best peptides from eight generations are shown: G1, G10, G25, G50, G75, G84, G96, and G100. Generations G84 and G96 were chosen because they presented the lowest overall values ​​for the experiments with ColabFold and Rosetta, respectively. The other generations were selected to illustrate how the algorithm evolved peptides.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e presents the top five results for the experiments using ColabFold and Rosetta.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eTop 5 results for ColabFold and Rosetta. REU\u0026thinsp;=\u0026thinsp;Rosetta Energy Unit, which was used as a docking score.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"8\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colspan=\"8\" nameend=\"c8\" namest=\"c1\"\u003e\u003cp\u003eColabFold\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e#\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eGeneration\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSequence\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eDocking score (REU)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eAvg. REU\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eWorst REU\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eTime (s)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eOccupancy\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e84\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eAGVAKAKAV\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-634\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-525\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e10,104\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e79%\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eVKAKKCVI\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-633\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-518\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-133\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e8,091\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e79%\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eVAKCCG\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-628\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-509\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e56\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e9,019\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e79%\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e63\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eAAAKKVTKH\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-625\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-521\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-87\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e7,974\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e79%\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eLAKFKIKH\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-621\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-516\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e44,824\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e77%\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"8\" nameend=\"c8\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003eRosetta\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e#\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eGeneration\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSequence\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eDocking score (REU)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eAvg. REU\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eWorst REU\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eTime (s)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eOccupancy\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePGGHSCC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-611\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-468\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e448\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e1,672\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e70%\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e36\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eGMLELHQTYT\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-608\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-478\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-189\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e1,765\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e70%\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e99\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eGSSSSSYGSGC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-606\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-495\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-456\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e1,693\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e67%\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e94\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eHQSHLSHGCL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-605\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-486\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-428\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e1,732\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e70%\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eTLILGTERELLESYI\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-602\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-470\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-271\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e1,616\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e70%\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eFor the experiment using ColabFold, the lowest score value was obtained for the peptide AGVAKAKAV obtained in generation 84: -634 REU. In the same generation, the average docking score was \u0026minus;\u0026thinsp;525 REU, with the worst result being \u0026minus;\u0026thinsp;96 REU. The occupancy rate was 79% (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). For the experiment with Rosetta, the lowest score value was obtained for the peptide PGGHSCC in generation 96 with a docking score of -611 REU and an occupancy of 70%. In the same generation, the average docking score value was \u0026minus;\u0026thinsp;468 REU, with the worst result being 448 REU. In all evaluated metrics for the top results, the Rosetta results were inferior to those of ColabFold (except for execution time). In this regard, it is important to highlight that the recorded times for Rosetta refer to the generation of 5 poses per peptide, a number far below the recommended number of poses for a comprehensive conformational search\u003csup\u003e38\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eWhen analyzing the structure of the protein-peptide complex (Mpro-AGVAKAKAV), we can see that it makes a series of contacts between the different chains, calculated using the COCαDA tool [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e] (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e5\u003c/span\u003e, Table\u0026nbsp;2). For example, we can cite the predicted hydrogen bonds between T26 (threonine 26 of the protein) and V9 (valine 9 of the peptide), or a salt bridge between E166 (glutamate 166 of the protein) and K7 (lysine 7 of the peptide).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003e​​Table\u0026nbsp;2.\u003c/b\u003e Interatomic contacts detected on the interface of the complex Mpro-AGVAKAKAV, calculated using the COCαDA tool [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e].\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"No\" id=\"Taba\" border=\"1\"\u003e\u003ccolgroup cols=\"9\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003e#\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eContact\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eChain1\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eR1\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eAtom1\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eChain2\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eR2\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c8\"\u003e\u003cp\u003eAtom2\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c9\"\u003e\u003cp\u003eDistance\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eT26/V9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eT26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eV9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eO\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e2.78\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eT26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eO\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eV9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e3.19\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eF140/K7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eF140\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eO\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eK7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eNZ\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e2.63\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eG143/K7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eG143\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eK7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eO\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e2.89\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eG143/A8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eG143\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eA8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eO\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e3.68\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eS144/K7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eS144\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eK7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eO\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e3.14\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003eC145/K7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eC145\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eK7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eO\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e3.13\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eC145\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eSG\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eK7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e3.07\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eC145\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eSG\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eK7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eO\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e3.55\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eE166/K5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eE166\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eK5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eO\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e2.87\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eE166\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eO\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eK5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e2.85\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eE166/K7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eE166\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eOE1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eK7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eNZ\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e2.83\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eE166\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eOE2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eK7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eNZ\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e2.9\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eQ189/A6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eQ189\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eOE1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eA6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e3.29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eT190/A4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eT190\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eO\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eA4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e2.89\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eQ192/G2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eQ192\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eG2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eO\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e3.73\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eAlso, when we consider all 100 generations, we can see that the protein-peptide binding energy of the models generated by ColabFold was lower for the most part (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e6\u003c/span\u003e). It is essential to note that the lower the binding energy, the stronger the binding force.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eHowever, we can see that, in the first generations, the models proposed by Rosetta had a lower score (for example, generation 1, 4, 5, 6, 8, 11). This suggests that after a certain number of generations, the strategy using ColabFold consistently outperformed Rosetta, as it appeared to model the proposed peptide within the binding site in a more consistent and complementary pose (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e7\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eRegarding the occupancy metric, the best results of each generation varied between 60% and 80%. The occupancy defines which original residues of the Mpro binding site are likely to interact with a residue of the peptide. In the case of the Rosetta experiment, the occupancy values ​​varied throughout the experiment. In the case of the ColabFold experiments, the values ​​increased over the generations. The maximum occupancy observed was 79% (Supplementary Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eWe also observed that the average size of the peptides with the best docking score decreased over the generations, as shown in the data from the experiment using Rosetta (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e8\u003c/span\u003e, above). For the experiment using ColabFold, the average peptide size was less than 10 amino acids in almost all generations. For over 40 generations, the best peptide had only five amino acids (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e8\u003c/span\u003e, below).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003eUsing manual curation to detect a better ligand\u003c/h2\u003e\u003cp\u003eThe results of the case study suggest that the AGVAKAKAV peptide is a potential binder for the Mpro protein. However, analysis of the peptide\u0026rsquo;s structure indicated that its size could be reduced by removing N-terminal residues, specifically alanine at position 1 and glycine at position 2. The literature has shown that shorter peptides tend to bind better to the receptor [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]. Wier \u0026amp; Beekman (2025) suggest that shortening the peptide sequence (truncation), allowing it to contain only residues essential for the interaction, can improve the efficiency of the ligand and may also simplify the synthesis process [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. Furthermore, alanine and glycine are small, neutral amino acids that generally contribute little to specific interactions with the protein's active site and may be dispensable for maintaining binding affinity and stability. Therefore, we hypothesized that the VAKAKAV peptide would bind more effectively to Mpro. To assess this, we performed molecular dynamics experiments to verify the binding of the Mpro-N3 complex (the original structure), the Mpro-AGVAKAKAV complex, and the Mpro-VAKAKAV complex. The molecular dynamics results are discussed in the next section.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003eMolecular dynamics simulations\u003c/h2\u003e\u003cp\u003eTo evaluate the structural stability and binding affinity of the complexes formed between SARS-CoV-2 Mpro and the proposed peptides, we performed 100-ns molecular dynamics simulations followed by RMSD, RMSF, hydrogen-bonding interactions, and free-energy calculations using the MM-PBSA method. These parameters enabled us to directly compare the dynamic behavior of the three defined systems (Mpro-AGVAKAKAV, Mpro-VAKAKAV, and the Mpro-N3 control) and relate it to the structural data previously described for the N3 (02JAVLPJE010) inhibitor in wild-type Mpro [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eRMSD analysis revealed distinct behaviors between the systems containing different types of ligands (Supplementary Figure S2). The Mpro-N3 system, in which Mpro is associated with a ligand (non-peptide chemical structure), exhibited sharper oscillations and abrupt fluctuations throughout the simulation, suggesting a less stable interaction with the catalytic site. In contrast, the complexes formed with the proposed peptides (Mpro-AGVAKAKAV and Mpro-VAKAKAV) achieved more consistent values ​​after the initial phase, indicating structural stabilization of the protein\u0026ndash;ligand complex. Among them, Mpro-VAKAKAV exhibited a lower and more uniform plateau compared to MproAGVAKAKAV, suggesting greater cohesion and better accommodation of the peptide in the active site.\u003c/p\u003e\u003cp\u003eRMSF analysis along the Mpro residues showed apparent differences between the systems (Supplementary Figure S3). The Mpro-N3 system, associated with the crystallographic ligand, consistently exhibited higher RMSF values across virtually the entire length of the protein, indicating greater residual mobility and less conformational restriction. In contrast, the complexes formed with the peptides (Mpro-AGVAKAKAV and Mpro-VAKAKAV) presented very similar profiles, with reduced RMSF values ​​across most residues, suggesting that the binding to the peptide promoted local stabilization of the protein. Peaks with greater fluctuation were observed in specific regions, possibly corresponding to flexible loops, but without compromising overall stability. These results indicate that the peptides contribute to stiffening of the active site and adjacent regions, reinforcing the stabilizing role already suggested by the RMSD analyses.\u003c/p\u003e\u003cp\u003eHydrogen bond interactions, as observed over the simulation, revealed differences between the systems (Supplementary Figure S4). Mpro-VAKAKAV shows a more persistent and continuously distributed hydrogen bond throughout the simulation, ensuring greater stability to the complex. Consistent interactions were observed with catalytic residues, including D153, C156, and D155, which remained throughout most of the simulation. Mpro-AGVAKAKAV also established important bonds, such as those involving E166, G143, and H164, but in a slightly less continuous manner compared to Complex 2. The Wild-type system (Mpro\u0026thinsp;+\u0026thinsp;N3) presented a more diffuse pattern, with less regular interactions, mostly displaced to regions peripheral to the catalytic site, such as V157 \u0026harr; K100 and S158 \u0026harr; N151, which do not fully correspond to the hotspots described as critical for Mpro inhibition \u003csup\u003e10\u003c/sup\u003e. These results reinforce that the proposed peptides favor greater stabilization of the active site compared to the N3 wild-ligand, with Mpro-VAKAKAV promoting the most consistent hydrogen bond interactions throughout the simulation.\u003c/p\u003e\u003cp\u003eHydrogen bond pairs throughout the simulation reinforce the differences between the systems (Fig.\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e9\u003c/span\u003e). Mpro-AGVAKAKAV complex also exhibited persistent bonds, E166 \u0026harr; K5 and K7 \u0026harr; E166, in addition to interactions with G143 and H164, all of which are important residues of the S1 hotspot, part of the substrate-recognition pocket of the Mpro [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. Peptides making interactions with such residues are thus highly likely to inhibit the pocket\u0026rsquo;s activity [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. Mpro-VAKAKAV complex demonstrated consistency, with several residue pairs maintaining high occupancy throughout virtually the entire 100-ns runtime. The interactions with D153, D55, and C156 are notable, remaining active throughout, and demonstrate a continuous and well-distributed contact network within the catalytic site. This behavior suggests that Mpro-VAKAKAV complex promoted a good stabilization of Mpro. The Mpro-N3 system presented a more irregular pattern, with diffusely distributed contacts and less temporal continuity. The main pairs (such as V157 \u0026harr; K100 and S158 \u0026harr; N151) exhibited intermediate occupancy, but without the identical important active site residues observed in the peptides. Thus, the heatmaps corroborate that the proposed peptides interact more stably with the Mpro.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe cumulative frequency analysis of hydrogen bonds over the 100 ns simulations revealed notable differences among the three systems evaluated (Supplementary Figure S5). Mpro-AGVAKAKAV complex also exhibited relevant interactions, such as K7 \u0026harr; E166, E166 \u0026harr; K5, and G143 \u0026harr; K7, which correspond to key residues of Mpro. Mpro-VAKAKAV complex presented the highest recurrence values, with emphasis on the interactions K7 \u0026harr; E166, K100 \u0026harr; D155, and D153 \u0026harr; C156, which remained at high frequency throughout practically the entire trajectory. This pattern suggests that the peptide fosters a highly stable contact network, thereby contributing to firm anchoring at the catalytic site. On the other hand, the Wild-type system (Mpro-N3) presented lower frequencies distributed in pairs less central to the catalytic site, notably V157 \u0026harr; K100 and S158 \u0026harr; N151. Although these interactions were consistent, they do not fully reproduce the classic hotspots of protease inhibition, suggesting a less optimized docking pattern. Thus, the frequency data reinforce that the proposed peptides, especially Mpro-VAKAKAV, establish a more robust and persistent hydrogen bond network compared to the reference ligand.\u003c/p\u003e\u003cp\u003eFree energy predicted using the MM-PBSA method revealed differences between the systems (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Mpro-VAKAKAV complex presented the most favorable value (\u0026minus;\u0026thinsp;23.97 kcal/mol), followed by Mpro-AGVAKAKAV complex (\u0026minus;\u0026thinsp;20.71 kcal/mol), while the Wild system exhibited considerably lower affinity (\u0026minus;\u0026thinsp;8.88 kcal/mol).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eBinding free energy (ΔGbind) estimated by MM-PBSA for the simulated Mpro systems in complexes with the proposed peptides and with the reference ligand (N3).\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"2\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFree Energy\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMM/PBSA (Kcal/mol)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMpro-AGVAKAKAV\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-20,71\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMpro-VAKAKAV\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-23,97\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMpro-N3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-8,88\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eThe stability observed in the RMSD plots indicated that both peptides conferred greater structural cohesion to Mpro compared to the crystallographic ligand N3, with Mpro-VAKAKAV standing out by reaching more stable plateaus. The RMSF results showed that the peptide complexes significantly reduced residual flexibility, especially in regions close to the catalytic site, suggesting conformational constraint induced by peptide binding. Hydrogen bond analyses complemented these findings: Mpro-VAKAKAV established a more persistent and distributed network of interactions throughout the simulation, including stable contacts with critical residues such as D153, D155, and C156. In contrast, Mpro-AGVAKAKAV complex maintained relevant bonds with E166, G143, and H164, albeit less consistently. The Wild system, in turn, exhibited less regular interactions, with most displaced to regions peripheral to the active site, which was reflected in the less favorable free energy value.\u003c/p\u003e\u003cp\u003eThus, by correlating all analyses, it is clear that the proposed peptides promote greater structural stability, reduce protease flexibility, and establish more robust interaction networks with catalytic residues. Among them, Mpro-VAKAKAV complex stands out as the most promising candidate, as it combines the lowest stable RMSD, reduced local fluctuations, greater hydrogen bond persistence, and the most favorable binding free energy. These results suggest that Mpro-VAKAKAV has greater potential for Mpro inhibition compared to Mpro-AGVAKAKAV and the reference complex.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003eLimitations\u003c/h2\u003e\u003cp\u003eDespite presenting promising results, this study has some limitations that must be acknowledged. First, the procedure for defining new populations within the genetic algorithm is computationally demanding, particularly when modeling complete protein-peptide complexes using ColabFold. In our experiments, running the genetic algorithm for 100 generations required more than 14 days, even on a server equipped with an Nvidia A100 80GB GPU, 786TB RAM, and an AMD Ryzen Threadripper PRO 5995WX 64-core CPU. On the other hand, the docking-based experiments using the same system were completed in just over one day.\u003c/p\u003e\u003cp\u003eSecond, the experiments indicated that the AGVAKAKAV peptide exhibited favorable binding to Mpro. However, manual inspection suggested that the two N-terminal residues contributed minimally to protein binding. Based on this observation, we proposed a shorter variant (VAKAKAV), which was subsequently supported by molecular dynamics simulations. This outcome highlights that, while the proposed method can generate promising candidate ligands, expert knowledge remains crucial for refining peptide design. Moreover, the algorithm could be further improved by incorporating the size, physicochemical properties, and chemical nature of amino acids, which could enhance the prioritization of residues contributing most to the interaction energy. Additionally, strategies such as systematically truncating nonessential residues at the peptide termini could be explored to optimize binding affinity and reduce peptide flexibility, ultimately leading to more potent and specific binders\u003c/p\u003e\u003cp\u003eFinally, it should be noted that the N3 peptide complexed with Mpro contains non-canonical amino acids. This feature may have influenced the molecular dynamics simulations and binding energy calculations, potentially introducing bias into the evaluation process.\u003c/p\u003e\u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn this study, we employed a genetic algorithm (GA) approach, integrated with docking-based scoring and AI 3D structure modeling, to design peptide binders targeting the main protease (Mpro). Our results indicate that the approach, called EvoPepFold, was successful in finding peptides with a higher interaction energy between the protein-peptide complex. The optimization process identified promising candidates, including AGVAKAKAV and VAKAKAV, which exhibited improved predicted binding affinities compared to the initial peptide sequence (N3). These results suggest that GA-driven strategies can effectively explore sequence space to identify peptides with favorable interaction patterns at the protein interface. However, some limitations must be acknowledged. The current approach relies exclusively on docking scores (Rosetta energy) as a proxy for binding affinity, which may not fully capture the dynamics and solvation effects present in real biological environments. Future work will address experimental assays to validate the computational predictions and explore alternative score functions to evaluate complexes. We are confident that the insights gained in this study will help drive innovation in peptide engineering, opening new paths for their rational design and therapeutic use.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eCompeting interests\u003c/h2\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003ch2\u003eAuthor contributions statement\u003c/h2\u003e\n\u003cp\u003eFC developed the scripts and performed the experiments. FC and DM write the manuscript. SCA performed the molecular dynamics simulations. LB, APA, RPL, SCA, LMS, and RCMM edit the manuscript. All authors read and approved the final version of this manuscript.\u003c/p\u003e\n\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eThis study was financed in part by the Coordena\u0026ccedil;\u0026atilde;o de Aperfei\u0026ccedil;oamento de Pessoal de N\u0026iacute;vel Superior \u0026ndash; Brasil (CAPES) \u0026ndash; Finance Code 001.\u003c/p\u003e\n\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\n\u003cp\u003eFC developed the scripts and performed the experiments. FC and DM write the manuscript. SCA performed the molecular dynamics simulations. LB, APA, RPL, SCA, LMS, and RCMM edit the manuscript. All authors read and approved the final version of this manuscript.\u003c/p\u003e\n\u003ch2\u003eAcknowledgement\u003c/h2\u003e\n\u003cp\u003eThe authors would like to thank the Brazilian agencies CAPES, CNPQ, and Fapemig. This study was financed in part by the Coordena\u0026ccedil;\u0026atilde;o de Aperfei\u0026ccedil;oamento de Pessoal de N\u0026iacute;vel Superior \u0026ndash; Brasil (CAPES) \u0026ndash; Finance Code 001.\u003c/p\u003e\n\u003ch2\u003eData Availability\u003c/h2\u003e\n\u003cp\u003eData and scripts are available at [https://github.com/LBS-UFMG/evopepfold](https:/github.com/LBS-UFMG/evopepfold) .\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eFrappier, V., Duran, M., Keating, A. E. \u0026amp; PixelDB Protein\u0026ndash;peptide complexes annotated with structural conservation of the peptide binding mode. \u003cem\u003eProtein Sci.\u003c/em\u003e \u003cb\u003e27\u003c/b\u003e, 276\u0026ndash;285 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAngelova, A., Drechsler, M., Garamus, V. M. \u0026amp; Angelov, B. Pep-Lipid Cubosomes and Vesicles Compartmentalized by Micelles from Self-Assembly of Multiple Neuroprotective Building Blocks Including a Large Peptide Hormone PACAP-DHA. \u003cem\u003eChemNanoMat\u003c/em\u003e 5, 1381\u0026ndash;1389 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNissan, N., Allen, M. C., Sabatino, D. \u0026amp; Biggar, K. K. Future Perspective: Harnessing the Power of Artificial Intelligence in the Generation of New Peptide Drugs. \u003cem\u003eBiomolecules\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e, 1303 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLee, A. C. L., Harris, J. L., Khanna, K. K. \u0026amp; Hong J.-H. A comprehensive review on current advances in peptide drug development and design. \u003cem\u003eInt. J. Mol. Sci.\u003c/em\u003e \u003cb\u003e20\u003c/b\u003e, 2383 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLau, J. L. \u0026amp; Dunn, M. K. Therapeutic peptides: Historical perspectives, current development trends, and future directions. \u003cem\u003eBioorg. Med. Chem.\u003c/em\u003e \u003cb\u003e26\u003c/b\u003e, 2700\u0026ndash;2707 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLau, J. L. \u0026amp; Dunn, M. K. Therapeutic peptides: Historical perspectives, current development trends, and future directions. \u003cem\u003eBioorg. Med. Chem.\u003c/em\u003e \u003cb\u003e26\u003c/b\u003e, 2700\u0026ndash;2707 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHashemi, Z. S. et al. In silico Approaches for the Design and Optimization of Interfering Peptides Against Protein\u0026ndash;Protein Interactions. \u003cem\u003eFront Mol. Biosci\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e, (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDiakou, I. et al. Novel computational pipelines in antiviral structure\u0026ndash;based drug design (Review). \u003cem\u003eBiomed. Rep.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e, 1\u0026ndash;5 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eIbrahim, M. et al. Why is the Omicron main protease of SARS-CoV-2 less stable than its wild-type counterpart? A crystallographic, biophysical, and theoretical study. \u003cem\u003ehLife\u003c/em\u003e \u003cb\u003e2\u003c/b\u003e, 419\u0026ndash;433 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSacco, M. D. et al. Structure and inhibition of the SARS-CoV-2 main protease reveal strategy for developing dual inhibitors against Mpro and cathepsin L. \u003cem\u003eSci. Adv.\u003c/em\u003e \u003cb\u003e6\u003c/b\u003e, eabe0751 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNissan, N., Allen, M. C., Sabatino, D. \u0026amp; Biggar, K. K. Future Perspective: Harnessing the Power of Artificial Intelligence in the Generation of New Peptide Drugs. \u003cem\u003eBiomolecules\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e, 1303 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYan, Y., Zhang, D., Zhou, P., Li, B. \u0026amp; Huang, S. Y. HDOCK: a web server for protein\u0026ndash;protein and protein\u0026ndash;DNA/RNA docking based on a hybrid strategy. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cb\u003e45\u003c/b\u003e, W365\u0026ndash;W373 (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYan, Y., Tao, H., He, J. \u0026amp; Huang, S. Y. The HDOCK server for integrated protein\u0026ndash;protein docking. \u003cem\u003eNat. Protoc.\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e, 1829\u0026ndash;1852 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhou, P., Jin, B., Li, H. \u0026amp; Huang, S. Y. HPEPDOCK: a web server for blind peptide\u0026ndash;protein docking based on a hierarchical algorithm. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cb\u003e46\u003c/b\u003e, W443\u0026ndash;W450 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRaveh, B., London, N., Zimmerman, L. \u0026amp; Schueler-Furman, O. Rosetta FlexPepDock ab-initio: simultaneous folding, docking and refinement of peptides onto their receptors. \u003cem\u003ePloS One\u003c/em\u003e. \u003cb\u003e6\u003c/b\u003e, e18934 (2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChaudhury, S. et al. Benchmarking and analysis of protein docking performance in Rosetta v3. 2. \u003cem\u003ePloS One\u003c/em\u003e. \u003cb\u003e6\u003c/b\u003e, e22477 (2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEvans, R. et al. Protein complex prediction with AlphaFold-Multimer. biorxiv 2021\u0026ndash;10 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOmidi, A., M\u0026oslash;ller, M. H., Malhis, N., Bui, J. M. \u0026amp; Gsponer, J. AlphaFold-Multimer accurately captures interactions and dynamics of intrinsically disordered protein regions. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e 121, e2406407121 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFjell, C. D., Jenssen, H., Cheung, W. A., Hancock, R. E. W. \u0026amp; Cherkasov, A. Optimization of antibacterial peptides by genetic algorithms and cheminformatics. \u003cem\u003eChem. Biol. Drug Des.\u003c/em\u003e \u003cb\u003e77\u003c/b\u003e, 48\u0026ndash;56 (2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBorkakoti, N. \u0026amp; Thornton, J. M. AlphaFold2 protein structure prediction: Implications for drug discovery. \u003cem\u003eCurr. Opin. Struct. Biol.\u003c/em\u003e \u003cb\u003e78\u003c/b\u003e, 102526 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRaveh, B., London, N., Zimmerman, L. \u0026amp; Schueler-Furman, O. Rosetta FlexPepDock ab-initio: Simultaneous Folding, Docking and Refinement of Peptides onto Their Receptors. \u003cem\u003ePLOS ONE\u003c/em\u003e. \u003cb\u003e6\u003c/b\u003e, e18934 (2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChaudhury, S. et al. Benchmarking and Analysis of Protein Docking Performance in Rosetta v3.2. \u003cem\u003ePLOS ONE\u003c/em\u003e. \u003cb\u003e6\u003c/b\u003e, e22477 (2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRaveh, B., London, N., Zimmerman, L. \u0026amp; Schueler-Furman, O. Rosetta FlexPepDock ab-initio: Simultaneous Folding, Docking and Refinement of Peptides onto Their Receptors. \u003cem\u003ePLOS ONE\u003c/em\u003e. \u003cb\u003e6\u003c/b\u003e, e18934 (2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBerman, H. M. et al. The Protein Data Bank. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cb\u003e28\u003c/b\u003e, 235\u0026ndash;242 (2000).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMartins, P. M. et al. Propedia: a database for protein\u0026ndash;peptide identification based on a hybrid clustering algorithm. \u003cem\u003eBMC Bioinform.\u003c/em\u003e \u003cb\u003e22\u003c/b\u003e, 1 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMartins, P. et al. Propedia v2.3: A novel representation approach for the peptide-protein interaction database using graph-based structural signatures. \u003cem\u003eFront Bioinforma\u003c/em\u003e \u003cb\u003e3\u003c/b\u003e, (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMirdita, M. et al. ColabFold: making protein folding accessible to all. \u003cem\u003eNat. Methods\u003c/em\u003e. \u003cb\u003e19\u003c/b\u003e, 679\u0026ndash;682 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAdolf-Bryfogle, J. \u0026amp; Jr, R. L. D. The PyRosetta Toolkit: A Graphical User Interface for the Rosetta Software Suite. \u003cem\u003ePLOS ONE\u003c/em\u003e. \u003cb\u003e8\u003c/b\u003e, e66856 (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAbraham, M. J. et al. High performance molecular simulations through multi-level parallelism from laptops to supercomputers. \u003cem\u003eSoftwareX 1\u003c/em\u003e. \u003cb\u003eGROMACS\u003c/b\u003e, 19\u0026ndash;25 (2015).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHuang, J. et al. CHARMM36m: an improved force field for folded and intrinsically disordered proteins. \u003cem\u003eNat. Methods\u003c/em\u003e. \u003cb\u003e14\u003c/b\u003e, 71\u0026ndash;73 (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMalde, A. K. et al. An automated force field topology builder (ATB) and repository: version 1.0. \u003cem\u003eJ. Chem. Theory Comput.\u003c/em\u003e \u003cb\u003e7\u003c/b\u003e, 4026\u0026ndash;4037 (2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMichaud-Agrawal, N., Denning, E. J., Woolf, T. B. \u0026amp; Beckstein, O. MDAnalysis: a toolkit for the analysis of molecular dynamics simulations. \u003cem\u003eJ. Comput. Chem.\u003c/em\u003e \u003cb\u003e32\u003c/b\u003e, 2319\u0026ndash;2327 (2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGowers, R. J. et al. MDAnalysis: a Python package for the rapid analysis of molecular dynamics simulations. (Los Alamos National Laboratory (LANL), Los Alamos, NM (United States), (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHunter, J. D. \u0026amp; Matplotlib A 2D graphics environment. \u003cem\u003eComput. Sci. Eng.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, 90\u0026ndash;95 (2007).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVald\u0026eacute;s-Tresanco, M. S., Vald\u0026eacute;s-Tresanco, M. E., Valiente, P. A. \u0026amp; Moreno, E. gmx_MMPBSA: a new tool to perform end-state free energy calculations with GROMACS. \u003cem\u003eJ. Chem. Theory Comput.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e, 6281\u0026ndash;6291 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLemos, R. P., Mariano, D., de Silveira, S., de Melo-Minardi, R. C. \u0026amp; A. \u0026amp; COCαDA-A Fast and Scalable Algorithm for Interatomic Contact Detection in Proteins Using Cα Distance Matrices. \u003cem\u003eFront. Bioinforma\u003c/em\u003e. \u003cb\u003e5\u003c/b\u003e, 1630078 (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLemos, R., Mariano, D., Silveira, S. \u0026amp; Melo-Minardi, R. COCαDA - Large-Scale Protein Interatomic Contact Cutoff Optimization by Cα Distance Matrices. \u003cb\u003e70\u003c/b\u003e (2024). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.5753/bsb.2024.245545\u003c/span\u003e\u003cspan address=\"10.5753/bsb.2024.245545\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMeng, E. C. et al. UCSF ChimeraX: Tools for structure building and analysis. \u003cem\u003eProtein Sci.\u003c/em\u003e \u003cb\u003e32\u003c/b\u003e, e4792 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003evan Wier, S. P. \u0026amp; Beekman, M. A. Peptide design to control protein\u0026ndash;protein interactions. (2025). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1039/D4CS00243A\u003c/span\u003e\u003cspan address=\"10.1039/D4CS00243A\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLeenheer, D., ten Dijke, P. \u0026amp; Hipolito, C. J. A current perspective on applications of macrocyclic-peptide-based high-affinity ligands. \u003cem\u003ePept. Sci.\u003c/em\u003e \u003cb\u003e106\u003c/b\u003e, 889\u0026ndash;900 (2016).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJin, Z. et al. Structure of Mpro from SARS-CoV-2 and discovery of its inhibitors. \u003cem\u003eNature\u003c/em\u003e \u003cb\u003e582\u003c/b\u003e, 289\u0026ndash;293 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVerma, N., Henderson, J. A. \u0026amp; Shen, J. Proton-Coupled Conformational Activation of SARS Coronavirus Main Proteases and Opportunity for Designing Small-Molecule Broad-Spectrum Targeted Covalent Inhibitors. \u003cem\u003eJ. Am. Chem. Soc.\u003c/em\u003e \u003cb\u003e142\u003c/b\u003e, 21883\u0026ndash;21890 (2020).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"protein-peptide interactions, COVID-19, Mpro","lastPublishedDoi":"10.21203/rs.3.rs-7706745/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7706745/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003ePeptide inhibitors represent a promising class of antiviral therapeutics, offering several advantages over traditional small-molecule drugs, including low toxicity, high specificity, and biocompatibility. However, rational and efficient design and optimization of inhibitor peptides remains a significant challenge to current methods. Here we show EvoPepFold, a genetic algorithm-based framework designed to generate inhibitory peptides. We evaluated EvoPepFold to design and optimize peptides targeting the SARS-CoV-2 main protease (M\u003csup\u003epro\u003c/sup\u003e). EvoPepFold was applied through two complementary strategies: molecular docking using the Rosetta suite, and peptide 3D modeling with ColabFold. The top candidates were further evaluated through molecular dynamics simulations to assess stability and interaction energy. Our results demonstrate that EvoPepFold successfully identified peptides with favorable binding affinities and stable protein-peptide interactions. These findings highlight the potential of evolutionary algorithms in guiding the rational design of peptide-based antivirals, contributing to ongoing efforts in peptide engineering for therapeutic applications.\u003c/p\u003e","manuscriptTitle":"EvoPepFold: A Hybrid Evolutionary and Structural Pipeline for AI- Guided Peptide Inhibitor Design Using AlphaFold and Rosetta","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-15 17:57:38","doi":"10.21203/rs.3.rs-7706745/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-08T05:36:31+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-07T01:39:20+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-06T06:10:05+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-04T03:45:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"154594296679719194813078705212894034116","date":"2025-10-02T11:34:16+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"65536031559991861122982632388641905276","date":"2025-10-02T08:27:02+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"274820983988757476888915630867507499181","date":"2025-10-02T02:47:00+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"132556066452707245698932716117192625587","date":"2025-09-30T05:01:42+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"15452508513947629324761600015600493031","date":"2025-09-30T04:11:40+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-30T03:57:36+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-09-29T13:24:01+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-26T13:52:27+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-26T09:43:19+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-09-24T19:54:57+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"bd5b36a4-78c7-4208-a654-5bab8544fc5c","owner":[],"postedDate":"October 15th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":55936459,"name":"Biological sciences/Biochemistry"},{"id":55936460,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":55936461,"name":"Biological sciences/Drug discovery"},{"id":55936462,"name":"Biological sciences/Structural biology"}],"tags":[],"updatedAt":"2025-12-01T16:11:51+00:00","versionOfRecord":{"articleIdentity":"rs-7706745","link":"https://doi.org/10.1038/s41598-025-28061-y","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-11-29 15:58:37","publishedOnDateReadable":"November 29th, 2025"},"versionCreatedAt":"2025-10-15 17:57:38","video":"","vorDoi":"10.1038/s41598-025-28061-y","vorDoiUrl":"https://doi.org/10.1038/s41598-025-28061-y","workflowStages":[]},"version":"v1","identity":"rs-7706745","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7706745","identity":"rs-7706745","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}