all

pdb2reaction all runs the workflow end-to-end from PDB/mmCIF/XYZ/GJF. mmCIF and oversized PDB inputs are converted to a safely reindexed internal PDB; coordinate outputs also include CIF with the original identifiers.

all runs in one of three modes, chosen by what you pass:

• Multi-structure MEP ([mode] all (mep)) — give ≥ 2 structures in reaction order. With -c, all first extracts active-site models; without it, the full supplied structures are used. It then runs GSM / DMF MEP search and optionally runs TSOPT + IRC / freq / DFT per reactive segment. For extracted PDB/mmCIF input, --refine-path --write-ref-merge additionally writes a coordinate composite for inspection.

• Single-structure scan-defined workflow ([mode] all (scan-lists)) — give one structure plus one or more --scan-lists/-s literals. One literal defines one stage; several tuples inside it are advanced concertedly. The stage endpoints form the ordered input series for the MEP step.

• TSOPT-only — give a single input and set --tsopt (no --scan-lists). all skips the MEP / merge stages, runs tsopt + EulerPC IRC on the active-site model (or the full input if extraction is skipped), and identifies the higher-energy endpoint as the reactant.

Note

The TSOPT-only reactant/product labels follow an energy-order presentation convention, not a chemically established reaction direction: the higher-energy IRC endpoint is presented as the reactant (on an exact energy tie the left endpoint is the reactant). The R/P labels, reactant_irc/product_irc filenames, barrier and delta are computed under this convention. The machine-readable summary records it explicitly under endpoint_assignment (policy = "higher_energy_endpoint_as_reactant", chemical_direction_known = false); read that field and do not infer chemical direction from the labels alone.

Important

Without --tsopt, the workflow produces TS candidates (highest-energy images from MEP search). Adding --tsopt refines them and performs terminal exact-PHVA validation. Numerical optimizer convergence and saddle order are reported separately. all proceeds to IRC only when optimization converged, terminal PHVA completed, and a negative reaction direction is available. A converged higher-order stationary point (n_imag > 1) may be followed by diagnostic IRC with an explicit warning, but it is not certified as a first-order TS. Actual optimizer non-convergence, no imaginary mode, failed/unavailable PHVA, or no valid negative root stops the pipeline after preserving the TS artifacts and before IRC. Always inspect the imaginary modes and IRC endpoint connectivity before mechanistic interpretation.

Examples

Working examples for GPP C6-methyltransferase BezA (Tsutsumi et al., Angew. Chem. Int. Ed. 2022, 61, e202111217) covering both multi-structure MEP and scan-based pipelines: examples/.

Command form:

pdb2reaction all -i INPUT1 [INPUT2 ...] [-c SUBSTRATE] [-b uma|orb|mace|aimnet2] [options]

Multi-structure MEP with TS + thermo + DFT:

# Multi-structure MEP with TS + thermo + DFT
pdb2reaction all -i 1.R.pdb 3.P.pdb -c 'SAM,GPP,MG' -l 'SAM:1,GPP:-3' \
    --tsopt --thermo --dft --out-dir ./result_mep

Single-structure staged scan (two stages):

# Single-structure staged scan (two stages)
pdb2reaction all -i 1.R.pdb -c 'SAM,GPP,MG' -l 'SAM:1,GPP:-3' \
    -s '[("CS1 SAM 320","GPP 321 C7",1.60)]' '[("GPP 321 H11","GLU 186 OE2",0.90)]' \
    --tsopt --thermo --out-dir ./result_scan

TSOPT-only validation of a single TS candidate:

# TSOPT-only validation of a single TS candidate
pdb2reaction all -i TS_candidate.pdb -c 'SAM,GPP,MG' -l 'SAM:1,GPP:-3' \
    --tsopt --thermo --dft

PDB / CIF / GJF companion files are generated automatically when reference templates are available; CIF is emitted for mmCIF/oversized-PDB bridge inputs. Control with --convert-files (on by default).

pdb2reaction all --help shows core options; pdb2reaction all --help-advanced shows the full option list.

Workflow

Full system(s) (PDB / mmCIF / XYZ / GJF)
  ├─ (optional) active-site extraction `extract` — requires PDB/mmCIF when `-c` is used
  │   └─ active-site cluster model(s)
  │       ├─ (optional) staged scan `scan` — single-structure workflows
  │       │   └─ ordered intermediates
  │       └─ MEP search `path-opt` (default) or `path-search` (recursive, `--refine-path`)
  │           └─ MEP trajectory `mep_trj.xyz` + energy diagrams
  └─ (optional) TS optimization + IRC `tsopt` → `irc`
      ├─ (optional) thermochemistry `freq`
      └─ (optional) single-point DFT `dft`

all runs the following stages in order. Stages 0 and 1 are automatic preprocessing; the rest fire based on the flags you pass.

0. Structure bridge and preflight (automatic) — mmCIF, oversized/nonstandard PDB, and PDB altloc input are converted once to a safely reindexed internal PDB; altloc is selected coherently per residue. For an ordinary PDB with blank element columns, all runs add-elem-info. Standalone fix-altloc is only needed when you want a cleaned PDB deliverable; standalone commands use the same bridge. Missing element data must still be repaired for an ordinary PDB or supplied as mmCIF _atom_site.type_symbol.

1. Active-site model extraction (when -c/--center is set) — accepts PDB/mmCIF paths, IDs/names, CHAIN:RESNAME, and CHAIN:RESNAME:RESSEQ. Per-input internal PDBs are saved under <out-dir>/_work/models/; bridge inputs also produce CIF companions.

2. Optional staged scan (single-input only) — each --scan-lists/-s literal is a list of (i, j, target_Å) tuples. Atom indices use the original input ordering, 1-based by default (pass --no-scan-one-based to interpret them as 0-based), and are remapped to the active-site model ordering. Three-field selectors like 'TYR,285,CA' are order-flexible; use positional CHAIN:RESNAME:RESSEQ[ICODE]:ATOM for repeated names or numbering. Stages run sequentially (stage 2 starts from stage 1’s result), and the stage endpoints become the ordered intermediates that feed the MEP step.

3. MEP search — by default runs single-pass path-opt; --refine-path switches to recursive path-search. Recursive refinement can improve a poor HEI but can also split a noisy/bad path into unnecessary segments and increase cost, so it is off by default. Segmentation is only a candidate mechanism until TS/frequency/IRC validation. Raw engine output stays under _work; mep.pdb, bridge-input mep.cif, mep_trj.xyz, and the diagram are promoted to the top level.

4. Full-system coordinate composite for inspection (optional) — with PDB/mmCIF input and -c/--center, --refine-path --write-ref-merge writes mep_w_ref.pdb; per-segment equivalents remain under _work/path_search/.

5. Per-segment post-processing (reactive segments only — bridge segments without bond changes are skipped):

   • --tsopt — TS optimization on each HEI active-site model. The terminal result separates optimization_status from saddle_validation. Numerical non-convergence, zero imaginary modes, failed/unavailable terminal PHVA, or inability to select a negative root stops after the TS geometry and result fields are registered and before IRC; frequencies and modes are present only when terminal PHVA completes. A numerically converged higher-order stationary point is retained and may continue through warning-labeled diagnostic IRC; this does not certify a first-order TS. For Hessian TS optimizers, the MEP’s energy-upwinding Cartesian tangent candidates are passed through a CPU/file cache to guide reaction-root identity (legacy paths without readable energies use normalized secants). Dimer does not consume --ref-mode, so the handoff/cache is marked not applicable. With --no-tsopt-from-mep-tan, TSOPT selects its initial root from the initial-structure Hessian modes. A continued result is followed by EulerPC IRC and IRC-endpoint re-optimization with --thresh-post (default baker). The endpoint optimization working directory is retained with --dump and otherwise removed after completion. Endpoint RFO uphill rejection is disabled by default; pass --reject-uphill to enable it for endpoint re-optimization only.

   • --thermo — freq on (R, TS, P) for vibrational + thermochemistry data and an MLIP Gibbs diagram.

   • --dft — single-point DFT on (R, TS, P) and a DFT diagram. With --thermo, a DFT//MLIP Gibbs diagram (DFT energies + MLIP thermal correction) is also produced.

   • Shared overrides: --opt-mode, --opt-mode-post, --flatten, --hessian-calc-mode, --tsopt-max-cycles, --tsopt-out-dir, --freq-*, --dft-*, --dft-engine (GPU-first by default). Frozen-boundary PHVA always uses the constrained rigid-mode treatment; it is unrelated to the MEP-derived --ref-mode. For Hessian evaluation modes see Hessian evaluation mode.

6. TSOPT-only mode (single input + --tsopt, no --scan-lists) — skips MEP / merge; runs tsopt + EulerPC IRC and generates the same energy diagrams plus optional freq / DFT outputs.

Outputs

The tree has three top-level zones: deliverables at the root, per-segment deliverables under segments/seg_NN/, and pipeline scratch under _work/ (safe to rm -rf once you have the results you need).

out_dir/   (default: ./result_all/)
├─ summary.log                 # Text summary (authored at the root)
├─ summary.json                # JSON results
├─ mep.pdb                     # Concatenated MEP path (promoted from the engine)
├─ mep.cif                     # Bridge inputs only; original identifiers restored
├─ mep_w_ref.pdb               # Coordinate composite for inspection (--write-ref-merge)
├─ mep_w_ref.cif               # Bridge-template companion (--write-ref-merge)
├─ mep_trj.xyz                 # Full MEP trajectory
├─ energy_diagram_MEP.png      # All-segment MEP barriers
├─ energy_diagram_*.png        # Aggregated post-processing diagrams (MLIP / Gibbs / DFT, with --tsopt etc.)
├─ segments/                   # Per-reactive-segment deliverables (bridge segments are skipped)
│  └─ seg_NN/                  # 2-digit index, e.g. seg_01, seg_02
│     ├─ reactant.{pdb,cif,xyz,gjf} # Canonical R/TS/P; bridge inputs write PDB + CIF
│     ├─ ts.{pdb,cif,xyz,gjf}
│     ├─ product.{pdb,cif,xyz,gjf}
│     ├─ ts/                   # TS optimization output + vibrational analysis (--tsopt)
│     ├─ irc/                  # IRC trajectories + plots (--tsopt)
│     ├─ freq/{R,TS,P}/        # frequencies_cm-1.txt + thermoanalysis.yaml (--thermo)
│     └─ dft/                  # DFT single-point results (--dft)
└─ _work/                      # Pipeline scratch (safe to delete)
   ├─ models/                  # Extracted active-site model PDBs (model_<input_stem>.pdb, when extraction runs)
   ├─ scan/                    # Staged scan results (with --scan-lists)
   ├─ add_elem_info/           # Preflight element-symbol fills
   └─ path_opt/                # Raw MEP-engine output (path_search/ with --refine-path)

In TSOPT-only mode (single input + --tsopt, no --scan-lists) there is no MEP stage: the optimized R/TS/P plus ts/, irc/, freq/, and dft/ land directly under segments/seg_01/, and the MEP work directory (_work/path_opt/) is absent.

Note

The canonical structures are segments/seg_NN/reactant.*, ts.*, product.* — cite these when reporting mechanisms. The ts/, irc/, freq/, and dft/ subdirectories inside the same seg_NN/ hold the per-stage working files (e.g. ts/vib/imag_*_trj.xyz, irc/*_trj.xyz) for debugging a single stage. The raw MEP-search engine output under _work/path_opt/ is scratch — the products you need (mep.pdb, bridge-input mep.cif, mep_trj.xyz, energy_diagram_MEP.png) are already promoted to the root.

At -v 2 the console summarizes active-site charge resolution, scan stages, MEP progress (GSM / DMF), and per-stage timing. Resolved configuration details are shown at -v 3; see Verbosity levels.

Plot file naming

Energy-diagram filenames encode method and scope:

File	Generated when	Content
energy_diagram_MEP.png	path-opt / path-search completes	All-segment MEP barriers (raw GSM / DMF values)
energy_diagram_MLIP.png	per-segment tsopt + IRC completes	R → TS → P (MLIP energy)
energy_diagram_G_MLIP.png	per-segment thermo completes	R → TS → P (MLIP Gibbs)
energy_diagram_DFT.png	per-segment DFT completes	R → TS → P (DFT energy)
energy_diagram_G_DFT_plus_MLIP.png	per-segment DFT + thermo	R → TS → P (DFT energy + MLIP thermal correction)
energy_diagram_MLIP_all.png / _G_MLIP_all.png / _DFT_all.png / _G_DFT_plus_MLIP_all.png	all segments aggregated (variants for MLIP / Gibbs / DFT / DFT//MLIP Gibbs)	Combined across all segments
irc_plot.png (per segments/seg_NN/irc/)	per-segment IRC completes	IRC profile (MLIP energy along the trajectory)
irc_plot_all.png	all segments aggregated	IRC profiles concatenated across segments

Reading summary.log

The header identifies the all entry route as MEP, Scan, or TS-only and prints the absolute root output directory and, for MEP / Scan, the absolute internal path module directory (TS-only reports -). Internal engine names such as path-opt / path-search remain machine-readable metadata rather than the user-facing pipeline mode.

The log is organized into numbered sections:

• [1] Global MEP overview — image / segment counts, MEP trajectory plot paths, aggregate MEP energy diagram.

• [2] Segment-level MEP summary (MLIP path) — per-segment barriers (ΔE‡), reaction energies (ΔE), bond-change summaries.

• [3] Per-segment post-processing (TSOPT / Thermo / DFT) — TS imaginary-frequency checks, IRC outputs, MLIP / thermo / DFT energy tables.

• [4] Energy diagrams (overview) — diagram tables for MEP / MLIP / Gibbs / DFT plus an optional cross-method summary.

• [5] Output directory structure — a compact tree of generated files with inline annotations.

Reading summary.json

Top-level keys: out_dir, n_images, n_segments (run metadata and counts); segments (per-segment entries with index, tag, kind, barrier_kcal, delta_kcal, bond_changes); energy_diagrams (optional payloads with labels, energies_kcal, energies_au, ylabel, image paths). summary.json intentionally omits the formatted tables and filesystem tree from summary.log.

CLI options

Defaults shown are used when the option is not specified. The full flag list is in the generated command reference; the tables below cover the options that need explanation.

Input expectations:

• Extraction enabled (-c/--center): inputs must be PDB / mmCIF so residues can be located.

• Extraction skipped: inputs may be PDB / mmCIF / XYZ / GJF.

• Multi-structure runs require ≥ 2 structures. For full input-file requirements (hydrogens, element columns, atom-order parity), see CLI Conventions.

Charge is resolved via the standard priority chain (see CLI Conventions: Charge specification). In all, the charge derivation from active-site model extraction (when -c is set) acts as an additional priority layer. Spin resolution: --multiplicity CLI → .gjf template → default 1. Always provide --ligand-charge/-l for non-standard substrates so the correct net charge propagates to scan / MEP / TSOPT / DFT.

Input / output

Option	Description	Default
-i, --input PATH...	Two or more full structures in reaction order (single input allowed only with --scan-lists/-s or --tsopt).	Required
--ref-pdb FILE	Reference PDB/mmCIF topology when -i provides XYZ/GJF coordinates.	None
-o, --out-dir PATH	Top-level output directory.	./result_all/
--convert-files / --no-convert-files	Global toggle for XYZ / TRJ → PDB / CIF / GJF companions when templates are available. CIF requires a bridged mmCIF/oversized-PDB topology.	True
--dump / --no-dump	Dump MEP (GSM / DMF) trajectories. An explicit parent toggle is forwarded to path-search / path-opt and scan / tsopt; when omitted, each child resolves its YAML/default. With --thermo, freq always retains the internal thermoanalysis.yaml channel even when --no-dump is supplied.	False
--config FILE	Base YAML applied first.	None
--show-config / --no-show-config	Print the effective YAML plus parent settings before execution. Child-governing parent options are null when omitted, because each child then resolves its own YAML/default.	False
--dry-run / --no-dry-run	Validate options and print the plan. With --center, run extraction in a temporary directory to validate derived charge and electron parity. No scan/MEP/TSOPT/freq/DFT stage runs, and no persistent output is produced.	False

Charge / spin

Option	Description	Default
-l, --ligand-charge TEXT	Net charge or per-resname mapping used when -q is omitted (PDB/mmCIF metadata or --ref-pdb).	None
-q, --charge INT	Explicit net system charge with highest priority. If it differs from the extracted/workflow-derived value, warn and use -q; omit it for automatic derivation.	None
-m, --multiplicity INT	Spin multiplicity forwarded to all downstream steps.	1

Extraction

Option	Description	Default
-c, --center TEXT	PDB/mmCIF path, IDs/names, CHAIN:RESNAME, or CHAIN:RESNAME:RESSEQ.	Required for extraction
-r, --radius FLOAT	Active-site model inclusion cutoff (Å). 0 disables radius-based expansion, leaving the -c and --selected-resn selections.	2.6
--radius-het2het FLOAT	Independent hetero–hetero cutoff (Å). 0 is internally nudged to 0.001 Å to avoid empty selections (same as standalone extract).	0.0
--include-h2o / --no-include-h2o	Include waters (HOH / WAT / TIP3 / SOL).	True
--exclude-backbone / --no-exclude-backbone	Remove backbone atoms on non-substrate amino acids.	False
--add-linkh / --no-add-linkh	Add cap hydrogens for severed bonds.	True
--selected-resn TEXT	Force-include using the same ID/name/chain-qualified selector forms as --center.	""
--modified-residue TEXT	Comma-separated residue names to treat as amino acids. NAME:charge adds or overrides the nominal charge for this extraction; bare NAME defaults to 0.	""
--freeze-links / --no-freeze-links	Freeze cap parents in active-site model PDBs.	True
--freeze-atoms TEXT	Comma-separated 1-based indices frozen in every stage. With extraction, indices refer to the original full input and are mapped to the active-site model; without extraction, they refer to the current input. Merged with --freeze-links and YAML geom.freeze_atoms.	None

Built-in amino-acid names follow force-field-normalized Amber/CHARMM conventions. Raw PDB CCD name collisions are not inferred automatically; use --modified-residue NAME:charge to state the intended nominal charge.

MEP search

Note

--max-cycles-gsm and --max-cycles-dmf bound the MEP stage only; leave them unset to let each stage use its own default. The single-stage opt and tsopt subcommands keep their own --max-cycles.

Option	Description	Default
--mep-mode [gsm|dmf]	MEP algorithm: GSM (Growing String Method) or DMF (Direct Max Flux).	gsm
--max-nodes INT	Movable internal images per GSM/DMF segment. Both engines retain two endpoints, so total images = max_nodes + 2.	20
--gsm-param [equi|energy]	GSM node parameterization after string growth. energy concentrates nodes in high-energy regions and may be tried when an equidistant path skips the reaction-coordinate region near the HEI; it does not identify a TS.	equi
--max-cycles-gsm INT	Maximum GSM string-optimizer cycles.	300
--max-cycles-dmf INT	Maximum DMF IPOPT iterations.	300
--climb / --no-climb	Enable climbing image for standard GSM segments (bridge segments always disable climbing).	True
--opt-mode [grad|hess]	Workflow preset (grad → L-BFGS / Dimer, hess → RFO / RS-P-RFO). Token-to-algorithm mapping depends on scope — see --opt-mode (subcommand-dependent) for the per-subcommand table; note that all’s pre-opt default (grad) differs from tsopt’s default (hess).	grad
--thresh TEXT	Convergence preset for single-structure optimizations and scan relaxations (gau_loose, gau, gau_tight, gau_vtight, baker, never).	gau
--thresh-gsm TEXT	Convergence preset for the GSM string optimizer of the MEP stage (same presets as --thresh).	gau_loose
--thresh-dmf TEXT	IPOPT dual-infeasibility tolerance of the DMF MEP stage: tight (0.04), middle (0.10), loose (0.20), or a positive float. Not a Gaussian preset.	tight
--preopt / --no-preopt	Pre-optimize active-site model endpoints before MEP search. Standalone scan / scan2d / scan3d default --preopt to False.	True
--refine-path / --no-refine-path	Enable recursive path-search with automatic bond-change segmentation / use the default single-pass path-opt per adjacent pair.	disabled
--write-ref-merge	Write mep_w_ref* / hei_w_ref* coordinate composites for inspection. Requires --refine-path, -c/--center, and PDB/mmCIF input.	disabled

MLIP calculator

Option	Description	Default
--workers, --workers-per-node	UMA predictor parallelism. workers > 1 cannot be combined with an explicit analytical Hessian request; use workers = 1 or finite differences. See workers > 1 is incompatible with analytical Hessians (UMA backend).	1, 1
--hessian-calc-mode [Analytical|FiniteDifference]	Shared MLIP Hessian engine.	FiniteDifference
-b, --backend {uma,orb,mace,aimnet2}	MLIP backend.	uma

Post-processing

Option	Description	Default
--tsopt / --no-tsopt	Run TS optimization + IRC per reactive segment.	False
--tsopt-from-mep-tan / --no-tsopt-from-mep-tan	For Hessian TS optimizers, guide reaction-root identity with CPU/file-cached HEI tangent candidates. Turning it off disables cache creation/use and selects from the initial Hessian modes. Not applicable to Dimer.	True
--thermo / --no-thermo	Run vibrational analysis (freq) on R / TS / P; requires --tsopt.	False
--dft / --no-dft	Run single-point DFT on R / TS / P; requires --tsopt.	False
--opt-mode-post [grad|hess]	Optimizer preset for TSOPT + post-IRC (grad → Dimer / L-BFGS, hess → RS-P-RFO / RFO).	hess
--thresh-post TEXT	Convergence preset for post-IRC endpoint optimizations.	baker
--flatten / --no-flatten	Enable surplus-imaginary-mode flattening in tsopt.	False
--reject-uphill / --no-reject-uphill	Opt in to rejecting energy-raising RFO steps during post-IRC endpoint re-optimization only, using a 1e-4 Hartree tolerance (roll back to the lower-energy geometry and shrink the trust radius); TS optimization forces rejection off, and path search is unaffected. At the emergency floor, the retained endpoint receives a final normal convergence check.	False
--irc-step-size FLOAT	Override the IRC maximum EulerPC step (Bohr). If IRC stops after only a few frames, retry with a smaller value such as 0.05.	IRC default 0.10
--irc-never-stop / --no-irc-never-stop	Ignore IRC gradient and energy stop conditions and trace each branch until its max-cycle cap. Numerical/integration failures and external interruption still stop propagation.	False

Warning

--dft cost and memory depend on basis-function count, elements, functional, grid, engine, and hardware; atom count alone is not a reliable cutoff. Pilot a representative state and monitor peak memory before selecting resources.

TSOPT optimizer selection order: --opt-mode-post (if set) → --opt-mode (only when explicitly provided) → TSOPT default (hess → rsprfo). Example: --opt-mode grad --opt-mode-post hess uses L-BFGS for path optimization and RS-P-RFO for TS refinement.

TSOPT / freq / DFT / scan overrides

Option	Description	Default
--tsopt-max-cycles INT	Override tsopt --max-cycles.	100000
--tsopt-out-dir PATH	Custom tsopt subdirectory.	None
--freq-out-dir PATH	Base directory override for freq outputs.	None
--freq-max-write INT	Maximum modes to write.	10
--freq-amplitude-ang FLOAT	Mode-trajectory amplitude (Å).	0.8
--freq-n-frames INT	Frames per mode trajectory.	20
--freq-sort [value|abs]	Mode sorting behavior.	value
--freq-temperature FLOAT	Thermochemistry temperature (K).	298.15
--freq-pressure FLOAT	Thermochemistry pressure (atm).	1.0
--dft-engine [gpu|cpu]	DFT backend (GPU4PySCF or PySCF). In all the option is named --dft-engine; the standalone dft subcommand uses --engine.	gpu
--dft-out-dir PATH	DFT outputs base directory override.	None
--dft-func-basis TEXT	Functional / basis pair.	wb97m-v/def2-tzvpd
--dft-max-cycle INT	Maximum SCF iterations.	100
--dft-conv-tol FLOAT	SCF convergence tolerance.	1e-9
--dft-grid-level INT	PySCF grid level.	3
-s, --scan-lists TEXT...	Staged scans: (i, j, target_Å) tuples (single-input runs).	None
--scan-out-dir PATH	Override the scan output directory.	None
--scan-one-based / --no-scan-one-based	How to read the --scan-lists atom indices: True = 1-based, False = 0-based.	None (1-based)
--scan-max-step-size FLOAT	Maximum step size (Å).	0.20
--scan-bias-k FLOAT	Harmonic bias strength (eV · Å⁻²).	300
--scan-relax-max-cycles INT	Relaxation max cycles per step.	100000
--scan-preopt / --no-scan-preopt	Override the scan preoptimization toggle.	None
--scan-endopt / --no-scan-endopt	Override the scan end-of-stage optimization toggle.	None

YAML configuration

all supports layered YAML — base settings via --config FILE, with the precedence defaults < config < CLI. The effective YAML is forwarded to every invoked subcommand, and each subcommand reads its own sections:

Subcommand	YAML sections
path-opt	geom, calc, gs, dmf, stopt, opt, lbfgs, rfo
path-search	geom, calc, gs, dmf, stopt, opt, lbfgs, rfo, bond, search
scan	geom, calc, opt, lbfgs, rfo, bias, bond
tsopt	geom, calc, opt, hessian_dimer, rsirfo
freq	geom, calc, freq, thermo
dft	dft
irc	geom, calc, irc

# Minimal example
calc:
  model: uma-s-1p2            # uma-s-1p2 | uma-m-1p1
  hessian_calc_mode: FiniteDifference   # default; benchmark Analytical before opting in
gs:
  max_nodes: 12
  climb: true
dft:
  grid_level: 6

Full schema: YAML Reference.

Notes

Structural differences outside the reaction coordinate can affect barriers obtained from independently prepared full-system structures. Inspect the structures and validate the selected path workflow for the modeled system.

• With --write-ref-merge, the static template is derived from the first original input; path-search --ref-full-pdb remains internal to this wrapper.

• Extraction radii: -r 0 (or --radius 0) disables radius-based expansion, so the model starts from residues selected by -c and --selected-resn; structural safeguards can still add a required disulfide partner or adjacent backbone context. The extractor internally clamps zero radii to 0.001 Å to avoid an empty geometric query.

• Energies in diagrams are reported relative to the first state (reactant) in kcal/mol.

• Omitting -c/--center skips extraction and feeds the entire input structures directly to MEP / tsopt / freq / dft; single-structure runs still require either --scan-lists/-s or --tsopt.

See Also

Installation · Getting Started · extract · scan · path-opt · path-search · tsopt · irc · freq · dft · Common Error Recipes · Troubleshooting · YAML Reference · Glossary.