Tomoyuki Miyao

Exhaustive Structure Generation for Inverse-QSPR/QSAR.

Chemical structure generation based on quantitative structure property relationship (QSPR) or quantitative structure activity relationship (QSAR) models is one of the central themes in the field of computer‐aided molecular design. The objective of structure generation is to find promising molecules, which according to statistical models, are considered to have desired properties. In this paper, a new method is proposed for the exhaustive generation of chemical structures based on inverse‐QSPR/QSAR. In this method, QSPR/QSAR models are constructed by multiple linear regression method, and then the conditional distribution of explanatory variables given the desired properties is estimated by inverse analysis of the models using the framework of a linear Gaussian model. Finally, chemical structures are exhaustively generated by a sophisticated algorithm that is based on a canonical construction path method. The usefulness of the proposed method is demonstrated using a dataset of the boiling points of acyclic hydrocarbons containing up to 12 carbon atoms. The QSPR model was constructed with 600 hydrocarbons and their boiling points. Using the proposed method, chemical structures which had boiling points of 100, 150, or 200 °C were exhaustively generated.

Inverse QSPR/QSAR Analysis for Chemical Structure Generation (from y to x)

Retrieving descriptor information (x information) from a value of an objective variable (y) is a fundamental problem in inverse quantitative structure-property relationship (inverse-QSPR) analysis but challenging because of the complexity of the preimage function. Herewith, we propose using a cluster-wise multiple linear regression (cMLR) model as a QSPR model for inverse-QSPR analysis. x information is acquired as a probability density function by combining cMLR and the prior distribution modeled with a mixture of Gaussians (GMMs). Three case studies were conducted to demonstrate various aspects of the potential of cMLR. It was found that the predictive power of cMLR was superior to that of MLR, especially for data with nonlinearity. Moreover, it turned out that the applicability domain could be considered since the posterior distribution inherits the prior distributions feature (i.e., training data feature) and represents the possibility of having the desired property. Finally, a series of inverse analyses with the GMMs/cMLR was demonstrated with the aim to generate de novo structures having specific aqueous solubility.

Systematic Generation of Chemical Structures for Rational Drug Design Based on QSAR Models

The first step in the process of drug development is to determine those lead compounds that demonstrate significant biological activity with regard to a target protein. Because this process is often costly and time consuming, there is a need to develop efficient methodologies for the generation of lead compounds for practical drug design. One promising approach for determining a potent lead compound is computational virtual screening. The biological activities of candidate structures found in virtual libraries are estimated by using quantitative structure activity relationship (QSAR) models and/or computational docking simulations. In virtual screening studies, databases of existing drugs or natural products are commonly used as a source of lead candidates. However, these databases are not sufficient for the purpose of finding lead candidates having novel scaffolds. Therefore, a method must be developed to generate novel molecular structures to indicate high activity for efficient lead discovery. In this paper, we review current trends in structure generation methods for drug design and discuss future directions. First, we present an overview of lead discovery and drug design, and then, we review structure generation methods. Here, the structure generation methods are classified on the basis of whether or not they employ QSAR models for generating structures. We conclude that the use of QSAR models for structure generation is an effective method for computational lead discovery. Finally, we discuss the problems regarding the applicability domain of QSAR models and future directions in this field.

Ring-System-Based Exhaustive Structure Generation for Inverse-QSPR/QSAR.

Inverse‐QSPR/QSAR aims to solve the inverse problem of chemical structure generation based on QSPR/QSAR models, once the properties or activities are specified. To efficiently solve this problem, an exhaustive ring‐system‐based structure generation methodology was developed. The concept of the applicability domain (AD) is automatically acknowledged within the proposed strategy. The local AD is considered by introducing the probability distribution of a given data set, and the universal AD is considered using ring‐system‐based fragments in the training data set. Structures with desired properties or activities are enumerated by assembling fragments, including atomic elements, in a tree‐like way. The usefulness of the proposed method is demonstrated through a case study of ligand design for the human alpha 2A adrenergic receptor (ADR2A_HUMAN). We succeeded in generating structures focusing only on a pre‐defined region in chemical space, resulting in structures whose desired activity has a high likelihood being efficiently generated. In addition, the limitations of our proposed method and future challenges are discussed.

Ring system-based chemical graph generation for de novo molecular design

Generating chemical graphs in silico by combining building blocks is important and fundamental in virtual combinatorial chemistry. A premise in this area is that generated structures should be irredundant as well as exhaustive. In this study, we develop structure generation algorithms regarding combining ring systems as well as atom fragments. The proposed algorithms consist of three parts. First, chemical structures are generated through a canonical construction path. During structure generation, ring systems can be treated as reduced graphs having fewer vertices than those in the original ones. Second, diversified structures are generated by a simple rule-based generation algorithm. Third, the number of structures to be generated can be estimated with adequate accuracy without actual exhaustive generation. The proposed algorithms were implemented in structure generator Molgilla. As a practical application, Molgilla generated chemical structures mimicking rosiglitazone in terms of a two dimensional pharmacophore pattern. The strength of the algorithms lies in simplicity and flexibility. Therefore, they may be applied to various computer programs regarding structure generation by combining building blocks.

Finding Chemical Structures Corresponding to a Set of Coordinates in Chemical Descriptor Space

When chemical structures are searched based on descriptor values, or descriptors are interpreted based on values, it is important that corresponding chemical structures actually exist. In order to consider the existence of chemical structures located in a specific region in the chemical space, we propose to search them inside training data domains (TDDs), which are dense areas of a training dataset in the chemical space. We investigated TDDs’ features using diverse and local datasets, assuming that GDB11 is the chemical universe. These two analyses showed that considering TDDs gives higher chance of finding chemical structures than a random search‐based method, and that novel chemical structures actually exist inside TDDs. In addition to those findings, we tested the hypothesis that chemical structures were distributed on the limited areas of chemical space. This hypothesis was confirmed by the fact that distances among chemical structures in several descriptor spaces were much shorter than those among randomly generated coordinates in the training data range

Computational method for estimating progression saturation of analog series

In lead optimization, it is difficult to estimate when an analog series might be saturated and synthesis of additional compounds would be unlikely to yield further progress. Rather than terminating a series, one often continues to generate analogs hoping to reach the final optimization goal, even if obstacles that are faced ultimately prove to be unsurmountable. Clearly, methodologies to better understand series progression and saturation are highly desirable. However, only a few approaches are currently available to monitor series progression and aid in decision making. Herein, we introduce a new computational method to assess progression saturation of an analog series by relating the properties of existing compounds to those of synthetic candidates and comparing their distributions in chemical space. The neighborhoods of analogs are analyzed and the distance relationships between existing and candidate compounds quantified. An intuitive dual scoring scheme makes it possible to characterize analog series and their degree of progression saturation.

Prediction of Compound Profiling Matrices Using Machine Learning

Screening of compound libraries against panels of targets yields profiling matrices. Such matrices typically contain structurally diverse screening compounds, large numbers of inactives, and small numbers of hits per assay. As such, they represent interesting and challenging test cases for computational screening and activity predictions. In this work, modeling of large compound profiling matrices was attempted that were extracted from publicly available screening data. Different machine learning methods including deep learning were compared and different prediction strategies explored. Prediction accuracy varied for assays with different numbers of active compounds, and alternative machine learning approaches often produced comparable results. Deep learning did not further increase the prediction accuracy of standard methods such as random forests or support vector machines. Target-based random forest models were prioritized and yielded successful predictions of active compounds for many assays.

Exploring differential evolution for inverse QSAR analysis

Inverse quantitative structure-activity relationship (QSAR) modeling encompasses the generation of compound structures from values of descriptors corresponding to high activity predicted with a given QSAR model. Structure generation proceeds from descriptor coordinates optimized for activity prediction. Herein, we concentrate on the first phase of the inverse QSAR process and introduce a new methodology for coordinate optimization, termed differential evolution (DE), that originated from computer science and engineering. Using simulation and compound activity data, we demonstrate that DE in combination with support vector regression (SVR) yields effective and robust predictions of optimized coordinates satisfying model constraints and requirements. For different compound activity classes, optimized coordinates are obtained that exclusively map to regions of high activity in feature space, represent novel positions for structure generation, and are chemically meaningful.

Identification of Bioactive Scaffolds Based on QSAR Models

In medicinal chemistry, the molecular scaffolds commonly found in compounds with preferable biological activities are called bioactive scaffolds. They are important because if present in a structure, it is more likely that the compound will be bioactive. Traditionally, medicinal chemists use their knowledge to identify bioactive scaffolds from a given data set after systematic extraction of all candidate scaffolds. However, manually sorting all the scaffolds is not practical as the number of compounds in a data set is often very large. Herein, we propose a method to systematically identify bioactive scaffolds based on a structure generator and a QSAR model. Two proof‐of‐concept studies showed that known bioactive scaffolds as well as scaffolds containing important substructures were extracted. The proposed method does not depend on scaffold frequencies in a data set, which is different from currently used methods for bioactive scaffold identification.