Query Database Schema¶
This page lists every field of the Query Database that can be referenced in a
query step’s filter expression (see Query Examples for
example syntax). This table is generated directly from the
chemreporter.query_database_tools.table_schemas
module.
Core Fields¶
key(String): Unique identifier for each structureentry_key(String): Unique identifier for each structuredatabase_name(String): Name of the source database (set when processing)split_name(String): Split name (set when processing)basis_set(String): Basis set used for the DFT calculationfunctional(String): Functional used for the DFT calculationcorrection_term(String): Dispersion or post-DFT correction termsubset(String): Dataset subset name (for example spice, anix, biomolecules)composition(String): Brutto formula of the structurenet_charge(Int64, e): Net molecular chargespin_multiplicity(Float64, unitless): Spin multiplicitynum_atoms(Int64): Number of atomsatomic_numbers(List(Int64)): Atomic numbers by atomenergy(Float64, eV): Raw DFT total energy (not cohesive/atomization energy)
Computed Properties¶
dipole_moment_magnitude(Float64, Debye): Magnitude of dipole momentnet_force_norm(Float64, eV/Å): Norm of net force vectormax_force_norm(Float64, eV/Å): Maximum force magnitude on any atomatomic_symbols(String): Unique element symbols in alphabetical ordermolecular_weight(Float64, g/mol): Molecular weightis_molecular_structure_valid(Boolean): Historic structure check (hydrogen coordination)num_water_molecules(Int64): Number of water moleculesgraph_properties_candidate(Boolean): Whether the structure is suitable for graph-based analysiserror_graph_properties(Boolean): Whether graph property calculation failedlogp(Float64, unitless): Partition coefficient (lipophilicity)tpsa(Float64, Ų): Topological polar surface areasmiles(String): SMILES representation
Molecular Fingerprints¶
fingerprint_0tofingerprint_1023(Int64, 0 or 1): 1024-bit Morgan fingerprint for similarity searches.
Bio motif counts¶
num_ala,num_arg,num_asn,num_asp,num_cys,num_gln,num_glu,num_gly,num_his,num_ile,num_leu,num_lys,num_met,num_phe,num_pro,num_ser,num_thr,num_trp,num_tyr,num_val(Int64): RDKit substructure counts per motif code.is_protein(Boolean): True when the sum of amino-acid motif counts exceeds 3num_ade,num_cyt,num_gua,num_thy,num_ura(Int64): RDKit substructure counts per motif code.is_nucleobase(Boolean): True when the sum of nucleobase motif counts exceeds 3
Reaction Properties¶
Present when extracted from OMOL25 source metadata for reactive subsets.
reaction_id(String): Reaction identifierreaction_pathway_id(Int64): Pathway identifier (if applicable)is_reactant(Boolean): Whether the structure is a reactantis_product(Boolean): Whether the structure is a productis_transition_state(Boolean): Whether the structure is a transition statereaction_step_idx(Int64): Index of the reaction step
Catalysis Properties¶
Present when extracted from OC20 supplementary metadata.
catalyst_bulk_id(String): Catalyst bulk identifiercatalyst_adsorbate_id(String): Catalyst adsorbate identifiercatalyst_bulk_symbols(String): Element symbols in the bulkcatalyst_adsorbate_symbols(String): Element symbols in the adsorbatecatalyst_adsorbate_smiles(String): SMILES of the adsorbatecatalyst_num_adsorbate_atoms(Int64): Number of atoms in the adsorbatecatalyst_num_bulk_atoms(Int64): Number of atoms in the bulkcatalyst_reference_energy(Float64, eV): Reference energy of the catalyst systemcatalyst_class(String): Catalyst class labelcatalyst_anomaly(String): Catalyst anomaly labelcatalyst_substrate_height(Float64, Å): Substrate heightcatalyst_miller_index(List(Int64)): Miller indices (h, k, l) of the surfacecatalyst_relaxation_frame_idx(Int64): Relaxation frame indexcatalyst_xyz_adsorbate_is_valid(Boolean): Whether adsorbate coordinates pass the OC20 xyz validity check
For more detailed information about the units used for each field, see Units and Physical Quantities.
Known caveats¶
num_water_molecules might be slightly overestimated: compounds containing an
R-OHH group are counted as containing one water molecule. Such cases are
assumed to be sufficiently rare, so this approximation is not currently
corrected for.