The landscape of small-molecule prodrugs

Prodrugs are derivatives with superior properties compared with the parent active pharmaceutical ingredient (API), which undergo biotransformation after administration to generate the API in situ. Although sharing this general characteristic, prodrugs encompass a wide range of different chemical structures, therapeutic indications and properties. Here we provide the first holistic analysis of the current landscape of approved prodrugs using cheminformatics and data science approaches to reveal trends in prodrug development. We highlight rationales that underlie prodrug design, their indications, mechanisms of API release, the chemistry of promoieties added to APIs to form prodrugs and the market impact of prodrugs. On the basis of this analysis, we discuss strengths and limitations of current prodrug approaches and suggest areas for future development.

Screening oral drugs for their interactions with the intestinal transportome via porcine tissue explants and machine learning

In vitro systems that accurately model in vivo conditions in the gastrointestinal tract may aid the development of oral drugs with greater bioavailability. Here we show that the interaction profiles between drugs and intestinal drug transporters can be obtained by modulating transporter expression in intact porcine tissue explants via the ultrasound-mediated delivery of small interfering RNAs and that the interaction profiles can be classified via a random forest model trained on the drug-transporter relationships. For 24 drugs with well-characterized drug-transporter interactions, the model achieved 100% concordance. For 28 clinical drugs and 22 investigational drugs, the model identified 58 unknown drug-transporter interactions, 7 of which (out of 8 tested) corresponded to drug-pharmacokinetic measurements in mice. We also validated the model's predictions for interactions between doxycycline and four drugs (warfarin, tacrolimus, digoxin and levetiracetam) through an ex vivo perfusion assay and the analysis of pharmacologic data from patients. Screening drugs for their interactions with the intestinal transportome via tissue explants and machine learning may help to expedite drug development and the evaluation of drug safety.

DeepDelta: predicting ADMET improvements of molecular derivatives with deep learning

Established molecular machine learning models process individual molecules as inputs to predict their biological, chemical, or physical properties. However, such algorithms require large datasets and have not been optimized to predict property differences between molecules, limiting their ability to learn from smaller datasets and to directly compare the anticipated properties of two molecules. Many drug and material development tasks would benefit from an algorithm that can directly compare two molecules to guide molecular optimization and prioritization, especially for tasks with limited available data. Here, we develop DeepDelta, a pairwise deep learning approach that processes two molecules simultaneously and learns to predict property differences between two molecules from small datasets. On 10 ADMET benchmark tasks, our DeepDelta approach significantly outperforms two established molecular machine learning algorithms, the directed message passing neural network (D-MPNN) ChemProp and Random Forest using radial fingerprints, for 70% of benchmarks in terms of Pearson's r, 60% of benchmarks in terms of mean absolute error (MAE), and all external test sets for both Pearson's r and MAE. We further analyze our performance and find that DeepDelta is particularly outperforming established approaches at predicting large differences in molecular properties and can perform scaffold hopping. Furthermore, we derive mathematically fundamental computational tests of our models based on mathematical invariants and show that compliance to these tests correlates with overall model performance - providing an innovative, unsupervised, and easily computable measure of expected model performance and applicability. Taken together, DeepDelta provides an accurate approach to predict molecular property differences by directly training on molecular pairs and their property differences to further support fidelity and transparency in molecular optimization for drug development and the chemical sciences.

Neuromuscular Development and Disease: Learning From in vitro and in vivo Models

The neuromuscular junction (NMJ) is a specialized cholinergic synaptic interface between a motor neuron and a skeletal muscle fiber that translates presynaptic electrical impulses into motor function. NMJ formation and maintenance require tightly regulated signaling and cellular communication among motor neurons, myogenic cells, and Schwann cells. Neuromuscular diseases (NMDs) can result in loss of NMJ function and motor input leading to paralysis or even death. Although small animal models have been instrumental in advancing our understanding of the NMJ structure and function, the complexities of studying this multi-tissue system in vivo and poor clinical outcomes of candidate therapies developed in small animal models has driven the need for in vitro models of functional human NMJ to complement animal studies. In this review, we discuss prevailing models of NMDs and highlight the current progress and ongoing challenges in developing human iPSC-derived (hiPSC) 3D cell culture models of functional NMJs. We first review in vivo development of motor neurons, skeletal muscle, Schwann cells, and the NMJ alongside current methods for directing the differentiation of relevant cell types from hiPSCs. We further compare the efficacy of modeling NMDs in animals and human cell culture systems in the context of five NMDs: amyotrophic lateral sclerosis, myasthenia gravis, Duchenne muscular dystrophy, myotonic dystrophy, and Pompe disease. Finally, we discuss further work necessary for hiPSC-derived NMJ models to function as effective personalized NMD platforms.