Miklos Feher

Property Distributions: Differences between Drugs, Natural Products, and Molecules from Combinatorial Chemistry

The differences between three different compound classes, natural products, molecules from combinatorial synthesis, and drug molecules, were investigated. The major structural differences between natural and combinatorial compounds originate mainly from properties introduced to make combinatorial synthesis more efficient. These include the number of chiral centers, the prevalence of aromatic rings, the introduction of complex ring systems, and the degree of the saturation of the molecule as well as the number and ratios of different heteroatoms. As drug molecules derive from both natural and synthetic sources, they cover a joint area in property space of natural and combinatorial compounds. A PCA-based scheme is presented that differentiates the three classes of compounds. It is suggested that by mimicking certain distribution properties of natural compounds, combinatorial products might be made that are substantially more diverse and have greater biological relevance.

A simple model for the prediction of blood-brain partitioning.

We derived a simple model for the prediction of blood-brain barrier penetration using three descriptors. The model contains the calculated octanol-water partition coefficient, the number of hydrogen-bond acceptors in an aqueous medium and the polar surface area. It was validated using an extensive dataset, comprising 100 diverse drug molecules. The descriptors are easily calculated and the model is suitable for the rapid prediction of the blood-brain barrier partitioning of drugs.

The use of consensus scoring in ligand-based virtual screening.

A new consensus approach has been developed for ligand-based virtual screening. It involves combining highly disparate properties in order to improve performance in virtual screening. The properties include structural, 2D pharmacophore and property-based fingerprints, scores derived using BCUT descriptors, and 3D pharmacophore approaches. Different approaches for the combination of all or some of these methods have been tested. Logistic regression and sum ranks were found to be the most advantageous in different pharmaceutical applications. The three major reasons consensus scoring appears to enrich data sets better than single scoring functions are (1) using multiple scoring functions is similar to repeated samplings, in which case the mean is closer to the true value than any single value, (2) due to the better clustering of actives, multiple sampling will recover more actives than inactives, and (3) different methods seem to agree more on the ranking of the actives than on the inactives. Furthermore, consensus results are not only better but are also more consistent across receptor systems.

Novel 2D fingerprints for ligand-based virtual screening

This paper describes the development of a set of new 2D fingerprints for the purposes of virtual screening in a pharmaceutical environment. The new fingerprints are based on established ones: the changes in their design included the introduction of overlapping pharmacophore feature types, feature counts for pharmacophore and structural fingerprints, as well as changes in the resolution in property description for property fingerprints. The effects of each of these changes on virtual screening performance were monitored using two types of training sets, emulating different stages in the drug discovery process. The results demonstrate that these changes all lead to an improvement in virtual screening performance.

Effect of Input Differences on the Results of Docking Calculations

The sensitivity of docking calculations to the geometry of the input ligand was studied. It was found that even small changes in the ligand input conformation can lead to large differences in the geometries and scores of the resulting docked poses. The accuracy of docked poses produced from different ligand input structures-the X-ray structure, the minimized Corina structure, and structures generated from conformational searches and molecular dynamics ensembles-were also assessed. It was found that using the X-ray ligand conformation as docking input does not always produce the most accurate docked pose when compared with other sources of ligand input conformations. Furthermore, no one method of conformer generation is guaranteed to always produce the most accurate docking pose. The docking scores are also highly sensitive to the source of the input conformation, which might introduce some noise in compound ranking and in binding affinity predictions. It is concluded that for the purposes of reproducibility and optimal performance, the most prudent procedure is to use multiple input structures for docking. The implications of these results on docking validation studies are discussed.

Numerical errors and chaotic behavior in docking simulations.

This work examines the sensitivity of docking programs to tiny changes in ligand input files. The results show that nearly identical ligand input structures can produce dramatically different top-scoring docked poses. Even changing the atom order in a ligand input file can produce significantly different poses and scores. In well-behaved cases the docking variations are small and follow a normal distribution around a central pose and score, but in many cases the variations are large and reflect wildly different top scores and binding modes. The docking variations are characterized by statistical methods, and the sensitivity of high-throughput and more precise docking methods are compared. The results demonstrate that part of docking variation is due to numerical sensitivity and potentially chaotic effects in current docking algorithms and not solely due to incomplete ligand conformation and pose searching. These results have major implications for the way docking is currently used for pose prediction, ranking, and virtual screening.

The Discovery of PLK4 Inhibitors: (E)-3-((1H-Indazol-6-yl)methylene)indolin-2-ones as Novel Antiproliferative Agents

The family of Polo-like kinases is important in the regulation of mitotic progression; this work keys on one member, namely Polo-like kinase 4 (PLK4). PLK4 has been identified as a candidate anticancer target which prompted a search for potent and selective inhibitors of PLK4. The body of the paper describes lead generation and optimization work which yielded nanomolar PLK4 inhibitors. Lead generation began with directed virtual screening, using a ligand-based focused library and a PLK4 homology model. Validated hits were used as starting points for the design and discovery of PLK4 inhibitors of novel structure, namely (E)-3-((1H-indazol-6-yl)methylene)indolin-2-ones. Computational models, based on a published X-ray structure (PLK4 kinase domain), were used to understand and optimize the in vitro activity of the series; potent antiproliferative activity was obtained. The kinase selectivity profile and cell cycle analysis of selected inhibitors are described. The results of a xenograft study with an optimized compound 50 (designated CFI-400437) support the potential of these novel PLK4 inhibitors for cancer therapy.

Rapid Prediction of Human Intestinal Absorption

This work describes the modeling of human intestinal absorption using one- and two-descriptor nonlinear models. Molecules with known transport mechanisms other than passive transport were removed from the dataset. The human intestinal absorption data were fitted to symmetric and asymmetric sigmoidal curves and surfaces that were based on the arctangent function. The descriptors employed in the models are the sum of hydrogen bond donors and acceptors and the calculated SlogP octanol-water partition coefficient. Despite the simplicity of the model, the quality of fits and predictions is comparable to more complex approaches. As descriptors and predicted values can be calculated rapidly, the method is ideally suited for qualitative predictions for large virtual libraries and the results from de novo design.

BHB: a simple knowledge-based scoring function to improve the efficiency of database screening.

A new knowledge-based scoring function was developed in this work to facilitate the rapid ranking of ligands in databases. The acronym of the method is BHB based on the descriptors it utilizes: buriedness, hydrogen bonding, and binding energy. Receptor buriedness is a measure of how well molecules occupy the binding pocket in comparison to known high-affinity ligands or, alternatively, whether they have contact with identified residues in the pocket. The possibility of hydrogen bond formation is checked for selected residues that are recognized as being important in the binding of known ligands. The approximate binding energy is calculated from the thermodynamic cycle using the optimized bound and free solvent conformations of the ligand-receptor system. The information necessary for the scoring function can ideally be gleaned from the 3D structure of the receptor-ligand complex. Alternatively, the descriptors can be derived from the 3D structure of the unbound receptor, provided this receptor has a known ligand that binds to the given site with nanomolar activity. We show that the new scoring functions provide up to 12 times improvement in enrichment compared to the popular commercial docking program GOLD.

Discovery of inhibitors of the mitotic kinase TTK based on N-(3-(3-sulfamoylphenyl)-1H-indazol-5-yl)-acetamides and carboxamides

TTK kinase was identified by in-house siRNA screen and pursued as a tractable, novel target for cancer treatment. A screening campaign and systematic optimization, supported by computer modeling led to an indazole core with key sulfamoylphenyl and acetamido moieties at positions 3 and 5, respectively, establishing a novel chemical class culminating in identification of 72 (CFI-400936). This potent inhibitor of TTK (IC50=3.6nM) demonstrated good activity in cell based assay and selectivity against a panel of human kinases. A co-complex TTK X-ray crystal structure and results of a xenograft study with TTK inhibitors from this class are described.