10.1021/ci700404c.s001
Xiang
S. Wang
Xiang
S.
Wang
Hao Tang
Hao
Tang
Alexander Golbraikh
Alexander
Golbraikh
Alexander Tropsha
Alexander
Tropsha
Combinatorial QSAR Modeling of Specificity and Subtype
Selectivity of Ligands Binding to Serotonin Receptors 5HT1E and 5HT1F
American Chemical Society
2008
data mining approaches
Combinatorial QSAR Modeling
5 HT 1E ligands
MolConnZ descriptors
5 HT 1F ligands
Automated Lazy Learning
set
R 2 values
Several descriptor types
PDSP Ki Database
receptor subtype selectivity
MHF
value activity data
QSAR models
MOE
FSG
5 HT 1E agonists
Scripps Research Institute
5 HT 1E subtype selectivity data
combinatorial QSAR modeling
PLS
model binding affinity
5 HT 1F ligand binding
k Nearest Neighbor
5 HT 1E
Molecular Hologram Fingerprints
TSRI
GPCR
Serotonin Receptors 5 HT 1E
2008-05-27 00:00:00
Journal contribution
https://acs.figshare.com/articles/journal_contribution/Combinatorial_QSAR_Modeling_of_Specificity_and_Subtype_Selectivity_of_Ligands_Binding_to_Serotonin_Receptors_5HT1E_and_5HT1F/2936992
The Quantitative Structure−Activity Relationship (QSAR)
approach has been applied to model binding affinity and receptor subtype
selectivity of human 5HT1E and 5HT1F receptor−ligands. The
experimental data were obtained from the PDSP Ki Database. Several
descriptor types and data-mining approaches have been used in the
context of combinatorial QSAR modeling. Data mining approaches included <i>k</i> Nearest Neighbor, Automated Lazy Learning (ALL), and PLS;
descriptor types included MolConnZ, MOE, DRAGON, Frequent Subgraphs
(FSG), and Molecular Hologram Fingerprints (MHFs). Highly predictive
QSAR models were generated for all three data sets (i.e., for ligands
of both receptor subtypes and for subtype selectivity), and different
individual techniques were proved best in each case. For real value
activity data available for 5HT1E and 5HT1F ligand binding, models
were characterized by leave-one-out cross-validated <i>R</i><sup>2</sup> (<i>q</i><sup>2</sup>) for the training sets
and predictive <i>R</i><sup>2</sup> values for the test
sets. The best models for 5HT1E ligands were obtained with the <i>k</i>NN approach combined with MolConnZ descriptors (<i>q</i><sup>2</sup> = 0.69, <i>R</i><sup>2</sup> = 0.92);
for 5HT1F ligands ALL QSAR method using MolConnZ descriptors gave
the best results (<i>R</i><sup>2</sup> = 0.92). Rigorously
validated classification models were also developed for the 5HT1E/5HT1F
subtype selectivity data set with high correct classification accuracy
for both training (CCR<sub>train</sub>= 0.88) and test (CCR<sub>test</sub> = 1.00) sets using <i>k</i>NN with MolConnZ descriptors.
The external predictive power of QSAR models was further validated
by virtual screening of The Scripps Research Institute (TSRI) screening
library to recover 5HT1E agonists and antagonists (not present in
the original PDSP data set) with high enrichment factors. The successful
development of externally predictive and interpretative QSAR models
affords further design and discovery of novel subtype specific GPCR
agents.