Over the past fifteen years 3D-QSAR models generated by extracting relevant information from molecular interaction fields (MIFs) have become a standard technique in medicinal chemistry. With the expiration of the Tripos patent covering both the CoMFA methodology and the related chemometric/statistical analysis, such methods are now in the public domain. Recently we proposed a GRID/GOLPE “consensus 3D-QSAR” approach complementing binding affinity and functional potency to gain insight into the differences between a good binder and a good agonist across a series of ligands with affinity to the α4β2 nicotinic receptor. The choice of the most favourable superposition pattern was based on the “consensus 3D-QSAR principle”: identification of a model yielding good and balanced fit of the GRID MIFs to both pIC50 and pEC50 data. This model provided valuable hints on the putative structural motifs which influence the degree of agonism, and thus also on the activation mechanism of nicotinic α4β2 receptors. The need for automated exploration of a large number of different superposition schemes prompted us to develop Open3DQSAR, a new software aimed at high-throughput model building and evaluation based on a scriptable interface. High performance is attained through the implementation of parallel algorithms for PLS regression and variable selection, allowing to exploit the full computational power of multiprocessor machines. Output is arranged to be human and machine-readable, as well as exportable to spreadsheets, Gnuplot, Grace, MOE, Maestro, PyMOL for graphical visualization of results. Open3DQSAR is available as pre-built binaries for mainstream platforms (Linux, Windows, Mac OS X, Solaris, IRIX) along with fully commented C source code. We believe that Open3DQSAR may become a valuable tool for pharmacophore exploration based on automated fitting of MIFs to pharmacological data. The wide range of MIF formats which may be imported (GRID, Tripos CoMFA/CoMSIA, QM ESP maps, plain text files), together with the open-source nature of the project, make Open3DQSAR a highly flexible chemometric toolbox customizable to one’s needs.
Open3DQSAR: a new open-source pharmacophore explorer based on chemometric analysis of molecular interaction fields
TOSCO, Paolo;
2009-01-01
Abstract
Over the past fifteen years 3D-QSAR models generated by extracting relevant information from molecular interaction fields (MIFs) have become a standard technique in medicinal chemistry. With the expiration of the Tripos patent covering both the CoMFA methodology and the related chemometric/statistical analysis, such methods are now in the public domain. Recently we proposed a GRID/GOLPE “consensus 3D-QSAR” approach complementing binding affinity and functional potency to gain insight into the differences between a good binder and a good agonist across a series of ligands with affinity to the α4β2 nicotinic receptor. The choice of the most favourable superposition pattern was based on the “consensus 3D-QSAR principle”: identification of a model yielding good and balanced fit of the GRID MIFs to both pIC50 and pEC50 data. This model provided valuable hints on the putative structural motifs which influence the degree of agonism, and thus also on the activation mechanism of nicotinic α4β2 receptors. The need for automated exploration of a large number of different superposition schemes prompted us to develop Open3DQSAR, a new software aimed at high-throughput model building and evaluation based on a scriptable interface. High performance is attained through the implementation of parallel algorithms for PLS regression and variable selection, allowing to exploit the full computational power of multiprocessor machines. Output is arranged to be human and machine-readable, as well as exportable to spreadsheets, Gnuplot, Grace, MOE, Maestro, PyMOL for graphical visualization of results. Open3DQSAR is available as pre-built binaries for mainstream platforms (Linux, Windows, Mac OS X, Solaris, IRIX) along with fully commented C source code. We believe that Open3DQSAR may become a valuable tool for pharmacophore exploration based on automated fitting of MIFs to pharmacological data. The wide range of MIF formats which may be imported (GRID, Tripos CoMFA/CoMSIA, QM ESP maps, plain text files), together with the open-source nature of the project, make Open3DQSAR a highly flexible chemometric toolbox customizable to one’s needs.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
