Matched Molecular Pair Analysis on Large Melting Point Datasets: A Big Data Perspective

Characterization of covalent inhibitors that disrupt the interaction between the tandem SH2 domains of SYK and FCER1G phospho-ITAM

RNA sequencing and genetic data support spleen tyrosine kinase (SYK) and high affinity immunoglobulin epsilon receptor subunit gamma (FCER1G) as putative targets to be modulated for Alzheimer's disease (AD) therapy. FCER1G is a component of Fc receptor complexes that contain an immunoreceptor tyrosine-based activation motif (ITAM). SYK interacts with the Fc receptor by binding to doubly phosphorylated ITAM (p-ITAM) via its two tandem SH2 domains (SYK-tSH2). Interaction of the FCER1G p-ITAM with SYK-tSH2 enables SYK activation via phosphorylation. Since SYK activation is reported to exacerbate AD pathology, we hypothesized that disruption of this interaction would be beneficial for AD patients. Herein, we developed biochemical and biophysical assays to enable the discovery of small molecules that perturb the interaction between the FCER1G p-ITAM and SYK-tSH2. We identified two distinct chemotypes using a high-throughput screen (HTS) and orthogonally assessed their binding. Both chemotypes covalently modify SYK-tSH2 and inhibit its interaction with FCER1G p-ITAM, however, these compounds lack selectivity and this limits their utility as chemical tools.

Characterization of covalent inhibitors that disrupt the interaction between the tandem SH2 domains of SYK and FCER1G phospho-ITAM

RNA sequencing and genetic data support spleen tyrosine kinase (SYK) and high affinity immunoglobulin epsilon receptor subunit gamma (FCER1G) as putative targets to be modulated for Alzheimer's disease (AD) therapy. FCER1G is a component of Fc receptor complexes that contain an immunoreceptor tyrosine-based activation motif (ITAM). SYK interacts with the Fc receptor by binding to doubly phosphorylated ITAM (p-ITAM) via its two tandem SH2 domains (SYK-tSH2). Interaction of the FCER1G p-ITAM with SYK-tSH2 enables SYK activation via phosphorylation. Since SYK activation is reported to exacerbate AD pathology, we hypothesized that disruption of this interaction would be beneficial for AD patients. Herein, we developed biochemical and biophysical assays to enable the discovery of small molecules that perturb the interaction between the FCER1G p-ITAM and SYK-tSH2. We identified two distinct chemotypes using a high-throughput screen (HTS) and orthogonally assessed their binding. Both chemotypes covalently modify SYK-tSH2 and inhibit its interaction with FCER1G p-ITAM.

Trends in small molecule drug properties: A developability molecule assessment perspective

Developability molecule assessment is a key interfacial capability across the biopharmaceutical industry, screening and staging molecules discovered by medicinal chemists for successful chemistry manufacturing controls (CMC) development and launch. The breadth of responsibility and expertise such teams possess puts them in a unique position to understand the impact of the physicochemical properties of a drug during its initial discovery and subsequent development. However, most of the publications describing trends in physicochemical properties are written from a medicinal chemistry perspective with the aim to identify molecules with better ADMET profiles that are either lead-like or drug-like, failing to describe the impact these properties have on CMC development. To systematically uncover knowledge obtained from recent trends in physicochemical properties and the corresponding impact on CMC development, a comprehensive analysis was conducted on molecules in the drug repurposing hub dataset. The only physicochemical property that seems to have been preserved in FDA-approved oral molecules over the decades (1900-2020) is a constant H-bond donor count, highlighting the importance this property has on cell permeability and lattice energy. Pharmaceutical attrition analysis suggests that partition-distribution coefficient, H-bond acceptors, polar surface area and the fraction of sp³ carbons are properties that are associated with compound attrition. Looking at pharmaceutical attrition asynchronously with the temporal analysis of FDA-approved oral molecules highlights the opposing trends, risks and diminishing effects some of these physiochemical properties (cLogP, cLogD and Fsp3) have on describing compound attrition during the past decade. Trellising the dataset by target class suggests that certain formulation and drug delivery strategies can be anticipated or put into place based on target class of a molecule. For example, molecules binding to nuclear hormone receptors are amenable to lipid-based drug delivery systems with proven commercial success. Although the poor solubility of kinase inhibitors is a combination of hydrophobicity (due to aromaticity) required to bind to its target and high lattice energy (melting point), they are a challenging target class to formulate. The influence of drug targets on physicochemical properties and the temporal nature of these properties is highlighted when comparing molecules in the drug repurposing dataset to those developed at Amgen. An improved understanding of the impact of molecular properties on performance attributes can accelerate decisions and facilitate risk assessments during candidate selection and development.

Physicochemical Profiling and Comparison of Research Antiplasmodials and Advanced Stage Antimalarials with Oral Drugs

To understand the property space of antimalarials, we collated a large dataset of research antiplasmodial (RAP) molecules with known in vitro potencies and advanced stage antimalarials (ASAMs) with established oral bioavailability. While RAP molecules are "non-druglike", ASAM molecules display properties closer to Lipinski's and Veber's thresholds. Comparison within the different potency groups of RAP molecules indicates that the in vitro potency is positively correlated to the molecular weight, the calculated octanol-water partition coefficient (clog P), aromatic ring counts (#Ar), and hydrogen bond acceptors. Despite both categories being bioavailable, the ASAM molecules are relatively larger and more lipophilic, have a lower polar surface area, and possess a higher count of heteroaromatic rings than oral drugs. Also, antimalarials are found to have a higher proportion of aromatic (#ArN) and basic nitrogen (#BaN) counts, features implicitly used in the design of antimalarial molecules but not well studied hitherto. We also propose using descriptors scaled by the sum of #ArN and #BaN (SBAN) to define an antimalarial property space. Together, these results may have important applications in the identification and optimization of future antimalarials.

QSAR-assisted-MMPA to expand chemical transformation space for lead optimization

Matched molecular pairs analysis (MMPA) has become a powerful tool for automatically and systematically identifying medicinal chemistry transformations from compound/property datasets. However, accurate determination of matched molecular pair (MMP) transformations largely depend on the size and quality of existing experimental data. Lack of high-quality experimental data heavily hampers the extraction of more effective medicinal chemistry knowledge. Here, we developed a new strategy called quantitative structure-activity relationship (QSAR)-assisted-MMPA to expand the number of chemical transformations and took the logD7.4 property endpoint as an example to demonstrate the reliability of the new method. A reliable logD7.4 consensus prediction model was firstly established, and its applicability domain was strictly assessed. By applying the reliable logD7.4 prediction model to screen two chemical databases, we obtained more high-quality logD7.4 data by defining a strict applicability domain threshold. Then, MMPA was performed on the predicted data and experimental data to derive more chemical rules. To validate the reliability of the chemical rules, we compared the magnitude and directionality of the property changes of the predicted rules with those of the measured rules. Then, we compared the novel chemical rules generated by our proposed approach with the published chemical rules, and found that the magnitude and directionality of the property changes were consistent, indicating that the proposed QSAR-assisted-MMPA approach has the potential to enrich the collection of rule types or even identify completely novel rules. Finally, we found that the number of the MMP rules derived from the experimental data could be amplified by the predicted data, which is helpful for us to analyze the medicinal chemical rules in local chemical environment. In summary, the proposed QSAR-assisted-MMPA approach could be regarded as a very promising strategy to expand the chemical transformation space for lead optimization, especially when no enough experimental data can support MMPA.

A machine learning workflow for molecular analysis: application to melting points

Computational tools encompassing integrated molecular prediction, analysis, and generation are key for molecular design in a variety of critical applications. In this work, we develop a workflow for molecular analysis (MOLAN) that integrates an ensemble of supervised and unsupervised machine learning techniques to analyze molecular data sets. The MOLAN workflow combines molecular featurization, clustering algorithms, uncertainty analysis, low-bias dataset construction, high-performance regression models, graph-based molecular embeddings and attribution, and a semi-supervised variational autoencoder based on the novel SELFIES representation to enable molecular design. We demonstrate the utility of the MOLAN workflow in the context of a challenging multi-molecule property prediction problem: the determination of melting points solely from single molecule structure. This application serves as a case study for how to employ the MOLAN workflow in the context of molecular property prediction.

Facts, Patterns, and Principles in Drug Discovery: Appraising the Rule of 5 with Measured Physicochemical Data

The rule of 5 was designed to estimate the likelihood of poor absorption or permeation, noting the impact of poor solubility. This Perspective explores the impact of various physicochemical descriptors and contemporary lipophilicity measurements on permeability and solubility, showing that the distribution coefficient log D_7.4 (rather than log P) is the most impactful parameter. Molecular weight, almost invariably the defining characteristic of "beyond the rule of 5" compounds, has little impact on solubility when log D_7.4 measurements and aromaticity are considered. Predicting permeation is more complex, given passive and carrier transport mechanisms; however, notable patterns of behavior are apparent, giving insight even "beyond the rule of 5". Recommended best practices should involve using the facts (measurements) and the patterns they reveal to establish informative principles rather than fastidious rules.

Spectral deep learning for prediction and prospective validation of functional groups

State-of-the-art identification of the functional groups present in an unknown chemical entity requires the expertise of a skilled spectroscopist to analyse and interpret Fourier transform infra-red (FTIR), mass spectroscopy (MS) and/or nuclear magnetic resonance (NMR) data. This process can be time-consuming and error-prone, especially for complex chemical entities that are poorly characterised in the literature, or inefficient to use with synthetic robots producing molecules at an accelerated rate. Herein, we introduce a fast, multi-label deep neural network for accurately identifying all the functional groups of unknown compounds using a combination of FTIR and MS spectra. We do not use any database, pre-established rules, procedures, or peak-matching methods. Our trained neural network reveals patterns typically used by human chemists to identify standard groups. Finally, we experimentally validated our neural network, trained on single compounds, to predict functional groups in compound mixtures. Our methodology showcases practical utility for future use in autonomous analytical detection.

Chemical space exploration guided by deep neural networks

A parametric t-SNE approach based on deep feed-forward neural networks was applied to the chemical space visualization problem. It is able to retain more information than certain dimensionality reduction techniques used for this purpose (principal component analysis (PCA), multidimensional scaling (MDS)). The applicability of this method to some chemical space navigation tasks (activity cliffs and activity landscapes identification) is discussed. We created a simple web tool to illustrate our work (http://space.syntelly.com).

A parametric t-SNE approach based on deep feed-forward neural networks was applied to the chemical space visualization problem.

Matched Molecular Pair Analysis on Large Melting Point Datasets: A Big Data Perspective

A matched molecular pair (MMP) analysis was used to examine the change in melting point (MP) between pairs of similar molecules in a set of ∼275k compounds. We found many cases in which the change in MP (ΔMP) of compounds correlates with changes in functional groups. In line with the results of a previous study, correlations between ΔMP and simple molecular descriptors, such as the number of hydrogen bond donors, were identified. In using a larger dataset, covering a wider chemical space and range of melting points, we observed that this method remains stable and scales well with larger datasets. This MMP-based method could find use as a simple privacy-preserving technique to analyze large proprietary databases and share findings between participating research groups.