Modern methods of drug discovery: an introduction

Prediction of drug-target interactions through multi-task learning

Identifying the binding between the target proteins and molecules is essential in drug discovery. The multi-task learning method has been introduced to facilitate knowledge sharing among tasks when the amount of information for each task is small. However, multi-task learning sometimes worsens the overall performance or generates a trade-off between individual task's performance. In this study, we propose a general multi-task learning scheme that not only increases the average performance but also minimizes individual performance degradation, through group selection and knowledge distillation. The groups are selected on the basis of chemical similarity between ligand sets of targets, and the similar targets in the same groups are trained together. During training, we apply knowledge distillation with teacher annealing. The multi-task learning models are guided by the predictions of the single-task learning models. This method results in higher average performance than that from single-task learning and classic multi-task learning. Further analysis reveals that multi-task learning is particularly effective for low performance tasks, and knowledge distillation helps the model avoid the degradation in individual task performance in multi-task learning.

Drug discovery and the use of computational approaches for infectious diseases

For centuries infectious diseases were the scourge of humanity, overcome only by the discovery of vaccination and penicillin. With an armamentarium of effective antibiotics, vaccines and drugs at hand, infectious diseases for many years were considered to be negligible. With the onset of the AIDS pandemic, the return of tuberculosis and influenza (e.g., swine influenza) this notion has changed in recent years. Drug discovery for infectious diseases, therefore, is again gaining increasing interest. This article discusses the drug-discovery process in this area and introduces major computational approaches used to identify suitable drug targets and to discover and optimize chemical lead compounds towards drug candidates using examples from antiparasitic drug discovery.

Comparison data sets for benchmarking QSAR methodologies in lead optimization

2D and 3D QSAR techniques are widely used in lead optimization-like processes. A compilation of 40 diverse data sets is described. It is proposed that these can be used as a common benchmark sample for comparisons of QSAR methodologies, primarily in terms of predictive ability. Use of this benchmark set will be useful for both assessment of new methods and for optimization of existing methods.

Systematic Statistical Comparison of Comparative Molecular Similarity Indices Analysis Molecular Fields for Computer-Aided Lead Optimization

Comparative molecular similarity indices analysis (CoMSIA) is a 3D quantitative structure-activity relationship technique used to determine structural and electronic features influencing biological activity. This proves particularly useful for facilitating lead optimization projects. This study aimed to compare CoMSIA models produced using different subsets of the CoMSIA molecular fields (steric, electrostatic, hydrophobic, hydrogen-bond donor, and hydrogen-bond acceptor) in a systematic and statistically valid manner. A total of 23 data sets sourced from the literature were used to compare molecular field contribution and model predictivity using leave-one-out cross-validated R2 values. Predictive ability varied in a highly statistically significant manner depending on the set of CoMSIA molecular fields used. In general, the greater the number of CoMSIA molecular fields included in the analysis, the better the model predictivity was. There is great redundancy in the information contained in the different CoMSIA molecular fields. When all five CoMSIA molecular fields are included, the hydrophobic and electrostatic fields had the largest and the steric field the smallest contribution. Data sets were clustered into four groups on the basis of the utility of molecular field sets to generate predictive models.

Utility of homology models in the drug discovery process

Advances in bioinformatics and protein modeling algorithms, in addition to the enormous increase in experimental protein structure information, have aided in the generation of databases that comprise homology models of a significant portion of known genomic protein sequences. Currently, 3D structure information can be generated for up to 56% of all known proteins. However, there is considerable controversy concerning the real value of homology models for drug design. This review provides an overview of the latest developments in this area and includes selected examples of successful applications of the homology modeling technique to pharmaceutically relevant questions. In addition, the strengths and limitations of the application of homology models during all phases of the drug discovery process are discussed.