Keywords in the Macro-pKa Input File

The keywords needed for the Macro-pKa input file (name.in) are listed in the table below.

Table 1. Macro-pKa input file keywords.

Keyword Type Allowed Values Default Values Description
active_atoms list of integers Any None Atom indices of tautomeric active sites.
add_stereoisomers boolean Any True Explicitly add stereoisomers involving pKa-active atoms when auto-generating tautomers.
autoconf_config string Any None Custom AutoConf config file for generating conformers and tautomers. These keywords are mutually exclusive: ['csrch', 'autoconf_config']
chiral_atoms list of integers Any None Atom indices for chiral centers to racemize.
conformer_files list of strings Any None .mae files containing precomputed tautomers and conformers. Note the "relative_charges" keyword still controls which charges are included in a calculation, regardless of the structures provided here. If no tautomers or conformers of a given charge are found in this file, the code will attempt to auto-generate them as usual.
csrch integer Any 0 Do conformational search with <N> confs per tautomer, where 0 implies conformational search is skipped. Note these keywords are mutually exclusive: ['csrch', 'autoconf_config']
debug boolean Any False Print extra debugging information.
deprotonate boolean Any False Deprotonate input structure.
infile string Any None Input .mae file containing one structure.
mlff_csrch boolean Any False Use MLFF to accelerate the filtering of conformers and tautomers.
ph real number Any 7.0 pH for tautomer populations.
pka_max real number Any 12.0 Maximum macro-pKa sweep cut-off value.
pka_min real number Any 2.0 Minimum macro-pKa sweep cut-off value.
protonate boolean Any False Protonate input structure.
qrnn_csrch boolean Any True Use QRNN to accelerate the filtering of conformers and tautomers.
relative_charges list of integers Any Automatic scan mode Included charge states relative to input structure.
split_stereoisomers boolean Any False Separate stereoisomers when calculating final populations.
use_ml boolean Any True Use Ligand-ML machine-learning method for the empirical corrections. If False, use Jaccard similarity method for empirical corrections.
prepare_input_structures boolean Any True Preprocess input structures, converting coordinates from 2D to 3D and adding any missing hydrogens.