Skip to content

Exploring the conformational and reactive dynamics of biomolecules in solution using an extended version of the glycine reactive force field

Evidence and attribution

Authority of statements

Prose below summarizes the publication identified by doi, title, and pdf_path. For definitive numerical values and schemes, use the peer-reviewed article.

Summary

This work extends an existing ReaxFF parameterization for glycine into a broader reactive model for peptides and proteins, aimed at protonation-state changes and reaction pathways involving amino acids in solution. The expansion adds a large quantum-mechanical training set (hundreds of systems) spanning all amino acids and selected short peptides. On sub-nanosecond trajectories (about 500 ps), ReaxFF predictions for representative biomolecular test cases—including capped amino acids, small peptides, and compact proteins—are compared to non-reactive classical simulations and to experimental benchmarks where available, including pharmaceutically motivated examples.

Methods

Grounding: papers/PCCP_Monti_ReaxFF_Peptides_accepted.pdf; normalized/extracts/2013pccp-venue-paper_p1-2.txt (abstract-level only).

1 — MD application (ReaxFF validation trajectories)

The abstract reports molecular dynamics validation of the extended ReaxFF protein model on a relatively short time scale (~500 ps) for well-characterized test cases including capped amino acids, peptides, and small proteins, including reaction mechanisms connected to the pharmaceutical sector, compared to classical non-reactive simulations and experimental reference data.

  • Engine / code: N/A — MD program and integrator are not named in the indexed excerpt; confirm in the full PDF (pdf_path).
  • System size & composition: Test systems are described qualitatively as capped amino acids, peptides, and small proteins (abstract); atom counts and stoichiometries are not in the p1–2 excerpt.
  • Boundaries / periodicity: N/A — PBC vs cluster details are not stated in the indexed excerpt.
  • Ensemble: N/A — NVE/NVT/NPT choice is not stated in the indexed excerpt.
  • Timestep: N/A — timestep (fs) is not stated in the indexed excerpt.
  • Duration / stages: ~500 ps validation trajectories are stated in the abstract for the performance assessment window.
  • Thermostat / barostat: N/A — thermostat/barostat types and coupling constants are not stated in the indexed excerpt.
  • Temperature: N/A — explicit temperature setpoints for the validation MD are not stated in the indexed excerpt.
  • Pressure: N/A — hydrostatic pressure control is not stated in the indexed excerpt.
  • Electric field: N/A — not used per abstract scope.
  • Replica / enhanced sampling: N/A — not stated in the indexed excerpt.

2 — Force-field training (ReaxFF extension for peptides/proteins)

  • Parent FF / elements: Reactive ReaxFF-based description framed as an expansion of previously reported glycine parameters for peptide/protein simulations (abstract).
  • QM reference: The training expansion is built from quantum mechanical calculations on >500 molecular systems (abstract); program, functional, basis, and k-sampling used for that QM reference set are not stated in the indexed excerpt—see the article Methods in pdf_path.
  • Training set: Adds >500 QM-characterized systems including all amino acids and some short peptide structures to the training set (abstract).
  • Optimization: N/A — fitting/optimization workflow and software are not stated in the indexed excerpt.
  • Reference data / validation: Post-training assessment compares ReaxFF predictions to non-reactive classical MD and experimental benchmarks on the ~500 ps validation window (abstract).

Findings

  • Outcomes & mechanisms: The abstract states that good agreement is obtained between ReaxFF-predicted conformations and kinetics and the reference classical simulations and experimental data for the highlighted test cases spanning capped amino acids, peptides, small proteins, and pharmaceutically motivated mechanisms.
  • Comparisons: Direct comparisons are framed against classical non-reactive MD and experiment on the same test cases (abstract).
  • Sensitivity / design levers: N/A — parameter sweeps (temperature, ionic strength, pH, etc.) are not summarized in the indexed excerpt beyond the qualitative validation framing.
  • Limitations & outlook: The abstract’s validation is explicitly on a short (~500 ps) time scale, which inherently limits claims about slow conformational transitions and rare reactive events without longer sampling in the primary text.
  • Corpus honesty: This page is grounded on an RSC accepted manuscript PDF and a short metadata-aligned extract; integrator, thermostat, system sizes, and QM training details must be taken from the full PDF when needed for reproduction.

Limitations

  • Short (hundreds of ps) trajectories do not resolve slow folding or rare conformational transitions.
  • Reactive FF accuracy is training-set dependent; transfer to uncommon chemistries or unusual protonation environments may require additional refinement.

Relevance to group

Foundational ReaxFF-for-biomolecules line from the van Duin group (with collaborators), connecting reactive hydrocarbon/organic frameworks to practical peptide and protein-scale simulations.

Citations and evidence anchors

  • DOI: 10.1039/C3CP51931G
  • Text-aligned pointer: normalized/extracts/2013pccp-venue-paper_p1-2.txt

Reader notes (navigation)

  • Corpus note: The ingested PDF is an RSC Accepted Manuscript; wording may differ slightly from the final edited issue PDF.
  • reaxff-family
  • Reactive models for organics and aqueous biomolecular chemistry