Viva Biotech's algorithm team publishes in JCIM: Pep2MARS eliminates the need for 'manual molecular Lego' in complex cyclic peptide molecular dynamics simulations

Recently, the Viva Biotech algorithm team publishedin the Journal of Chemical Information and Modeling(JCIM) an innovative research article titled"Pep2MARS: Automated Cyclic Peptide Parameterization for Molecular Dynamics and Compound Design".The paper was co-authored byDr. Yue Qian, Vice President of Viva Biotech Shanghai and Head of the MARS Business Unit.served as the corresponding author, along with the teamDr. Junhao Li, Senior Project Managerco-authored as the first author.

(Source: ACS journal website)

This study addresses the long-standing challenge of parameterization in pre-processing complex cyclic peptide systems for molecular dynamics (MD) simulations by proposing an automated modeling strategy based on 'Block'-level chemical units. It incorporates inter-residue covalent coupling relationships into the force field parameter generation framework, enabling a unified and stable representation of non-canonical amino acids, complex side-chain crosslinks, and polycyclic backbones. Further validation using microsecond-scale MD simulations of a PCSK9-targeting cyclic peptide revealed that peptide activity is not solely dependent on conformational stability in the unbound state, but is closely linked to conformational adaptability during binding—offering a new dynamic framework for understanding structure–activity relationships in complex cyclic peptides. This work not only enhances the standardization and reproducibility of complex cyclic peptide simulations but also provides critical computational infrastructure for high-fidelity mechanistic validation following AI-generated sequences, structure prediction, and virtual screening, thereby accelerating the large-scale implementation of cyclic peptide drug design pipelines.Pep2MARS fully integrates physics-based and AI-driven algorithmic frameworks and is committed to improving the success rate of complex molecule development through iterative optimization between computational modeling and wet-lab experiments.Below is the algorithm team’s further elaboration on this methodology.

The team has developed an automated parameterization workflow tailored forcomplex cyclic peptide systems: Starting from a three-dimensional structure—whether derived from in silico design or wet-lab experiments—it automatically identifies non-standard residues and covalent linkages and generates topology and parameter files compatible with the Amber force field. This transforms the pre-processing of MD simulations for complex cyclic peptides from an experience-dependent manual task into a standardized, reusable, batch-processable, and traceable workflow.

The greatest limitation of the traditional single-residue parameterization approach is its forced application of linear peptide modeling logic to cyclic peptides, often causing force field charges to overlook cross-residue chemical couplings introduced by cyclization.Viva Biotech’s proprietary workflow introduces a 'Block'-based modeling concept. By systematically classifying residues into single-residue, di-residue, and multi-residue blocks according to their connectivity patterns, this approach uniformly handles diverse cyclization strategies—including sidechain–sidechain, sidechain–backbone, backbone–backbone linkages—and fits atomic charges while optimizing force field parameters accordingly. This strategy transforms the parameterization process for complex cyclic peptides in molecular dynamics simulations from an experience-dependent, manual operation into an automated, chemistry-rule-based workflow that is verifiable and reproducible. The workflow has been successfully validated on ten complex PCSK9 cyclic peptide inhibitor systems: it not only automatically and robustly identifies cyclization relationships within these intricate systems and correctly generates parameter files, but also yields long-timescale MD trajectories consistent with those reported in the literature using other commercial software packages.

(Reprinted with permission from Li, J. and Qian, Y. (2026). Pep2MARS: Automated Cyclic Peptide Parameterization for Molecular Dynamics and Compound Design. Journal of Chemical Information and Modeling. https://doi.org/10.1021/acs.jcim.6c00340. Copyright 2026 American Chemical Society.)

With the rapid advancement of AI-driven molecular generation, structure prediction, and virtual screening capabilities, drug discovery is entering a new era of high-speed iterative 'design–validate' cycles. However, for complex cyclic peptide systems, achieving highly reliable automated simulation remains a critical bottleneck limiting the scalability of closed-loop design workflows.The significance of Pep2MARS lies not only in resolving parameterization challenges for complex systems but also in integrating structural modeling, parameter generation, and dynamic analysis into a reusable computational framework,providing essential infrastructure for cyclic peptide drug discovery in the AI era.

This workflow has already been deployed as a dedicated module within Viva Biotech’s official AIDD platform, primarily assisting internal users in efficiently preparing parameterization setups for complex cyclic peptides and related systems.

Going forward, Viva Biotech will continue advancing its computational drug discovery platform by integrating physics-based modeling with artificial intelligence, further strengthening synergy among complex molecule design, dynamic mechanism analysis, and experimental validation—to accelerate both innovative drug development efficiency and the discovery of complex molecular therapeutics.

For more details about the paper, please visit:

Li, J. and Qian, Y. (2026). Pep2MARS: Automated Cyclic Peptide Parameterization for Molecular Dynamics and Compound Design. Journal of Chemical Information and Modeling. doi:https://doi.org/10.1021/acs.jcim.6c00340.

Related recommendations

Viva Biotech (01873.HK), founded in 2008, provides one-stop comprehensive services to global innovative drug development enterprises, ranging from early structure-based drug discovery to commercial drug production. Leveraging our leading position in structure-based drug discovery (SBDD) technologies, we offer CRO services during the new drug research phase to global partners, utilizing advanced platforms such as X-ray protein crystallography, cryo-electron microscopy (Cryo-EM), DNA-encoded library technology (DEL), affinity mass spectrometry screening (ASMS), surface plasmon resonance (SPR), hydrogen-deuterium exchange mass spectrometry (HDX-MS), and AIDD/CADD. Our team, led by senior medicinal chemists and drug discovery biology experts, provides services including drug design, medicinal chemistry (H2L, LO), compound synthesis, chemical analysis and purification, kilogram-scale amplification, peptide synthesis, and corresponding bioactivity testing. Through our subsidiary, Longhua Pharmaceutical, we deliver end-to-end CMC/CDMO solutions from preclinical development to commercial production. Additionally, we focus on discovering and investing in high-potential biopharmaceutical startups, using a unique equity-for-service (EFS) business model to address unmet clinical needs.

$VIVA BIOTECH (01873.HK)$

13K Views