User Guide

Uploading QM logs

Drop a Gaussian, ORCA, or Psi4 log; get a semantic triple graph and ML-ready features in seconds.

Supported formats#

Gaussian 09 / 16 — .log / .out
ORCA 5 / 6 — .out
Psi4, NWChem, Q-Chem, CP2K — .out
Single file or ZIP archive (max 1 GB per file)

Uploading via UI#

Visit QM → Upload log, drag & drop a file, and check the validation toggles. Default options generate OntoCompChem RDF triples, semantic OntoSpecies mapping, and ML inputs.

Uploading via SDK#

python

from pymolhub import MolHub
mh = MolHub()

qm = mh.qm.parse(
    file="aspirin_b3lyp.log",
    generate_triples=True,
    validate_ontology=True,
)
print(qm.homo_eV, qm.lumo_eV, qm.band_gap_eV)

Extracted properties#

SCF / DFT energy
HOMO / LUMO orbital energies, band gap
Mulliken · CHELPG · NPA atomic charges
Dipole moment, vibrational frequencies, IR intensities
Enthalpy, entropy, free energy
Atom coordinates, bond lengths, angles

Semantic validation#

Each parse runs through OntoCompChem + OntoSpecies validation. If the knowledge base entails ψ (the parsed molecule's claim), the result is marked validated. Failed validations are flagged but not auto-rejected.

Validation failures usually point to missing convergence keywords (Gaussian) or missing Final Single Point Energy line (ORCA).

First docking job Exporting datasets