7wx7v-LigandScoringProtocol.txt

Name

SILCS-MC

Software

SILCS, SILCS-MC, MolCal

Parameters

Assumed pH 7.4

Method

Each atom in the ligand is assigned a classification type
to match its chemical nature to the FragMap types.
eg. A carbon atom in an aromatic ring will be assigned
the benzene FragMap type, and an alcohol oxygen will be assigned
methanol polar oxygen FragMap type. The atoms are then assigned
a score that equals the value of the grid free energy (GFE)
of the corresponding FragMap, which can be obtained for a single
conformation of the ligand or from an average over multiple
conformations. As the GFE values are both favorable (negative) and
unfavorable (positive) the per-atom GFE scores can be either
negative or positive, indicating the contribution of that particular
atom to the overall binding affinity. The sum of the per-atom GFE
scores gives the ligand GFE (LGFE) score following the appropriate
normalization. see Raman et al. J. Chem. Inf. Model. (2013), 53,
3384.

LGFE scores were computed by performing SILCS-MC simulations of the
ligands starting from a random orientation within a 5 A radius
sphere from the binding site. Binding site is determined from
previously available holo crystal structures. A SILCS-MC simulation
involves Monte-Carlo sampling of the ligand in translational,
rotational and torsional space, in the "field" of the GFE FragMaps.
The energy function is the LGFE, which is the normalized sum of per-atom
GFEs. The ligand doesn't "see" the protein, but the presence of the
exclusion map prevents it from going where the protein was rigid
during the SILCS simulations. This way, rapid sampling of the ligand
is obtained very efficiently, while at the same time including
protein flexibility in a mean-field like fashion. For more details
see Raman et al. J. Chem. Inf. Model. (2013), 53, 3384

Poses are predicted from the conformation withe lowest LGFE score.

7wx7v-PosePredictionProtocol.txt

Name

MC-SILCS

Software

SILCS, Molcal

System Preparation Parameters

Apo form of the structure was subjected to GCMC/MD SILCS.

System Preparation Method

Site Identification by Ligand Competitive Saturation (SILCS) was used to initially predict functional group free energy patterns at the FXR apo structure. Functional group free energy patterns are obtained in the form of Grid Free Energy (GFE) FragMaps with probes representing different chemical functionalities.

Pose Prediction Parameters

MC-SILCS sampling within 5 A within the binding pocket

Pose Prediction Method

Each atom in the ligand is assigned a classification type
to match its chemical nature to the FragMap types.
eg. A carbon atom in an aromatic ring will be assigned
the benzene FragMap type, and an alcohol oxygen will be assigned
methanol polar oxygen FragMap type. The atoms are then assigned
a score that equals the value of the grid free energy (GFE)
of the corresponding FragMap, which can be obtained for a single
conformation of the ligand or from an average over multiple
conformations. As the GFE values are both favorable (negative) and
unfavorable (positive) the per-atom GFE scores can be either
negative or positive, indicating the contribution of that particular
atom to the overall binding affinity. The sum of the per-atom GFE
scores gives the ligand GFE (LGFE) score following the appropriate
normalization. see Raman et al. J. Chem. Inf. Model. (2013), 53,
3384.

LGFE scores were computed by performing SILCS-MC simulations of the
ligands starting from a random orientation within a 5 A radius
sphere from the binding site. Binding site is determined from
previously available holo crystal structures. A SILCS-MC simulation
involves Monte-Carlo sampling of the ligand in translational,
rotational and torsional space, in the "field" of the GFE FragMaps.
The energy function is the LGFE, which is the normalized sum of per-atom
GFEs. The ligand doesn't "see" the protein, but the presence of the
exclusion map prevents it from going where the protein was rigid
during the SILCS simulations. This way, rapid sampling of the ligand
is obtained very efficiently, while at the same time including
protein flexibility in a mean-field like fashion. For more details
see Raman et al. J. Chem. Inf. Model. (2013), 53, 3384

Poses are predicted from the conformation withe lowest LGFE score.