BRUTUS: Optimization of a grid-based similarity function for rigid-body molecular superposition. II. Description and characterization

Virtual Screening Meets Deep Learning

BACKGROUND

Automated compound testing is currently the de facto standard method for drug screening, but it has not brought the great increase in the number of new drugs that was expected. Computer- aided compounds search, known as Virtual Screening, has shown the benefits to this field as a complement or even alternative to the robotic drug discovery. There are different methods and approaches to address this problem and most of them are often included in one of the main screening strategies. Machine learning, however, has established itself as a virtual screening methodology in its own right and it may grow in popularity with the new trends on artificial intelligence.

OBJECTIVE

This paper will attempt to provide a comprehensive and structured review that collects the most important proposals made so far in this area of research. Particular attention is given to some recent developments carried out in the machine learning field: the deep learning approach, which is pointed out as a future key player in the virtual screening landscape.

Centmitor-1, a novel acridinyl-acetohydrazide, possesses similar molecular interaction field and antimitotic cellular phenotype as rigosertib, on 01910.Na

Mitosis is an attractive target for the development of new anticancer drugs. In a search for novel mitotic inhibitors, we virtually screened for low molecular weight compounds that would possess similar steric and electrostatic features, but different chemical structure than rigosertib (ON 01910.Na), a putative inhibitor of phosphoinositide 3-kinase (PI3K) and polo-like kinase 1 (Plk1) pathways. Highest scoring hit compounds were tested in cell-based assays for their ability to induce mitotic arrest. We identified a novel acridinyl-acetohydrazide, here named as Centmitor-1 (Cent-1), that possesses highly similar molecular interaction field as rigosertib. In cells, Cent-1 phenocopied the cellular effects of rigosertib and caused mitotic arrest characterized by chromosome alignment defects, multipolar spindles, centrosome fragmentation, and activated spindle assembly checkpoint. We compared the effects of Cent-1 and rigosertib on microtubules and found that both compounds modulated microtubule plus-ends and reduced microtubule dynamics. Also, mitotic spindle forces were affected by the compounds as tension across sister kinetochores was reduced in mitotic cells. Our results showed that both Cent-1 and rigosertib target processes that occur during mitosis as they had immediate antimitotic effects when added to cells during mitosis. Analysis of Plk1 activity in cells using a Förster resonance energy transfer (FRET)-based assay indicated that neither compound affected the activity of the kinase. Taken together, these findings suggest that Cent-1 and rigosertib elicit their antimitotic effects by targeting mitotic processes without impairment of Plk1 kinase activity.

Integrating structure-based and ligand-based approaches for computational drug design

Methods utilized in computer-aided drug design can be classified into two major categories: structure based and ligand based, using information on the structure of the protein or on the biological and physicochemical properties of bound ligands, respectively. In recent years there has been a trend towards integrating these two methods in order to enhance the reliability and efficiency of computer-aided drug-design approaches by combining information from both the ligand and the protein. This trend resulted in a variety of methods that include: pseudoreceptor methods, pharmacophore methods, fingerprint methods and approaches integrating docking with similarity-based methods. In this article, we will describe the concepts behind each method and selected applications.

A novel, customizable and optimizable parameter method using spherical harmonics for molecular shape similarity comparisons

A novel molecular shape similarity comparison method, namely SHeMS, derived from spherical harmonic (SH) expansion, is presented in this study. Through weight optimization using genetic algorithms for a customized reference set, the optimal combination of weights for the translationally and rotationally invariant (TRI) SH shape descriptor, which can specifically and effectively distinguish overall and detailed shape features according to the molecular surface, is obtained for each molecule. This method features two key aspects: firstly, the SH expansion coefficients from different bands are weighted to calculate similarity, leading to a distinct contribution of overall and detailed features to the final score, and thus can be better tailored for each specific system under consideration. Secondly, the reference set for optimization can be totally configured by the user, which produces great flexibility, allowing system-specific and customized comparisons. The directory of useful decoys (DUD) database was adopted to validate and test our method, and principal component analysis (PCA) reveals that SH descriptors for shape comparison preserve sufficient information to separate actives from decoys. The results of virtual screening indicate that the proposed method based on optimal SH descriptor weight combinations represents a great improvement in performance over original SH (OSH) and ultra-fast shape recognition (USR) methods, and is comparable to many other popular methods. Through combining efficient shape similarity comparison with SH expansion method, and other aspects such as chemical and pharmacophore features, SHeMS can play a significant role in this field and can be applied practically to virtual screening by means of similarity comparison with 3D shapes of known active compounds or the binding pockets of target proteins.

Training a scoring function for the alignment of small molecules

A comprehensive data set of aligned ligands with highly similar binding pockets from the Protein Data Bank has been built. Based on this data set, a scoring function for recognizing good alignment poses for small molecules has been developed. This function is based on atoms and hydrogen-bond projected features. The concept is simply that atoms and features of a similar type (hydrogen-bond acceptors/donors and hydrophobic) tend to occupy the same space in a binding pocket and atoms of incompatible types often tend to avoid the same space. Comparison with some recently published results of small molecule alignments shows that the current scoring function can lead to performance better than those of several existing methods.

Spatial chemical distance based on atomic property fields

Similarity of compound chemical structures often leads to close pharmacological profiles, including binding to the same protein targets. The opposite, however, is not always true, as distinct chemical scaffolds can exhibit similar pharmacology as well. Therefore, relying on chemical similarity to known binders in search for novel chemicals targeting the same protein artificially narrows down the results and makes lead hopping impossible. In this study we attempt to design a compound similarity/distance measure that better captures structural aspects of their pharmacology and molecular interactions. The measure is based on our recently published method for compound spatial alignment with atomic property fields as a generalized 3D pharmacophoric potential. We optimized contributions of different atomic properties for better discrimination of compound pairs with the same pharmacology from those with different pharmacology using Partial Least Squares regression. Our proposed similarity measure was then tested for its ability to discriminate pharmacologically similar pairs from decoys on a large diverse dataset of 115 protein–ligand complexes. Compared to 2D Tanimoto and Shape Tanimoto approaches, our new approach led to improvement in the area under the receiver operating characteristic curve values in 66 and 58% of domains respectively. The improvement was particularly high for the previously problematic cases (weak performance of the 2D Tanimoto and Shape Tanimoto measures) with original AUC values below 0.8. In fact for these cases we obtained improvement in 86% of domains compare to 2D Tanimoto measure and 85% compare to Shape Tanimoto measure. The proposed spatial chemical distance measure can be used in virtual ligand screening.

Insights into ligand-elicited activation of human constitutive androstane receptor based on novel agonists and three-dimensional quantitative structure-activity relationship

The human constitutive androstane receptor (CAR, NR1I3) is an important regulator of xenobiotic metabolism and other physiological processes. So far, only few CAR agonists are known and no explicit mechanism has been proposed for their action. Thus, we aimed to generate a 3D QSAR model that could explain the molecular determinants of CAR agonist action. To obtain a sufficient number of agonists that cover a wide range of activity, we applied a virtual screening approach using both structure- and ligand-based methods. We identified 27 novel human CAR agonists on which a 3D QSAR model was generated. The model, complemented by coregulator recruitment and mutagenesis results, suggests a potential activation mechanism for human CAR and may serve to predict potential activation of CAR for compounds emerging from drug development projects or for chemicals undergoing toxicological risk assessment.

An efficient and accurate molecular alignment and docking technique using ab initio quality scoring

An accurate and efficient molecular alignment technique is presented based on first principle electronic structure calculations. This new scheme maximizes quantum similarity matrices in the relative orientation of the molecules and uses Fourier transform techniques for two purposes. First, building up the numerical representation of true ab initio electronic densities and their Coulomb potentials is accelerated by the previously described Fourier transform Coulomb method. Second, the Fourier convolution technique is applied for accelerating optimizations in the translational coordinates. In order to avoid any interpolation error, the necessary analytical formulas are derived for the transformation of the ab initio wavefunctions in rotational coordinates. The results of our first implementation for a small test set are analyzed in detail and compared with published results of the literature. A new way of refinement of existing shape based alignments is also proposed by using Fourier convolutions of ab initio or other approximate electron densities. This new alignment technique is generally applicable for overlap, Coulomb, kinetic energy, etc., quantum similarity measures and can be extended to a genuine docking solution with ab initio scoring.

Symmetry operation measures

We introduce a new mathematical tool for quantifying the symmetry contents of molecular structures: the Symmetry Operation Measures. In this approach, we measure the minimal distance between a given structure and the structure which is obtained after applying a selected symmetry operation on it. If the given operation is a true symmetry operation for the structure, this distance is zero; otherwise it gives an indication of how different the transformed structure is from the original one. Specifically, we provide analytical solutions for measures of all the improper rotations, S n p, including mirror symmetry and inversion, as well as for all pure rotations, C n p. These measures provide information complementary to the Continuous Symmetry Measures (CSM) that evaluate the distance between a given structure and the nearest structure which belongs to a selected symmetry point-group.