Augmenting a training dataset of the generative diffusion model for molecular docking with artificial binding pockets.

This study introduces the PocketCFDM generative diffusion model, aimed at improving the prediction of small molecule poses in the protein binding pockets. The model utilizes a novel data augmentation technique, involving the creation of numerous artificial binding pockets that mimic the statistical patterns of non-bond interactions found in actual protein-ligand complexes. An algorithmic method was developed to assess and replicate these interaction patterns in the artificial binding pockets built around small molecule conformers. It is shown that the integration of artificial binding pockets into the training process significantly enhanced the model's performance. Notably, PocketCFDM surpassed DiffDock in terms of non-bond interaction and steric clash numbers, and the inference speed. Future developments and optimizations of the model are discussed. The inference code and final model weights of PocketCFDM are accessible publicly via the GitHub repository: https://github.com/vtarasv/pocket-cfdm.git.

3DProtDTA: a deep learning model for drug-target affinity prediction based on residue-level protein graphs.

Accurate prediction of the drug-target affinity (DTA) in silico is of critical importance for modern drug discovery. Computational methods of DTA prediction, applied in the early stages of drug development, are able to speed it up and cut its cost significantly. A wide range of approaches based on machine learning were recently proposed for DTA assessment. The most promising of them are based on deep learning techniques and graph neural networks to encode molecular structures. The recent breakthrough in protein structure prediction made by AlphaFold made an unprecedented amount of proteins without experimentally defined structures accessible for computational DTA prediction. In this work, we propose a new deep learning DTA model 3DProtDTA, which utilises AlphaFold structure predictions in conjunction with the graph representation of proteins. The model is superior to its rivals on common benchmarking datasets and has potential for further improvement.

Organelle specific fluorescent phenomics and transcriptomic profiling to evaluate cellular response to tris(1,3 dichloro 2 propyl)phosphate.

Tris(1,3-dichloro-2-propyl)phosphate (TDCPP) has been suspected to cause toxicity invertebrates, but its phenotypic effects and the underlying regulatory mechanism have not been fully revealed. Generally, cellular responses tightly control and affect various phenotypes. The scope of the whole organism or cellular toxicological phenotyping, however, has been limited, and quantitative analysis methods using phenotype data have not been fully established. Here, we demonstrated that fluorescence imaging of sub-organelle-based phenomic analysis together with transcriptomic profiling can enable a comprehensive understanding of correlations between molecular and phenomic events. To reveal the cellular response to TDCPP exposure, we obtained three sub-organelle images as fluorescent phenotypes. Transcriptomic perturbation data were measured from the RNA-seq experiment, and both profiling results were analyzed together. Interestingly, organelle phenomic data showed a unique fluorescent intensity increase in the endoplasmic reticulum (ER), and pathway analysis using transcriptomic data also revealed that ER was significantly enriched in gene ontology terms. Following the series of analyses, RNA-seq data also revealed potential carcinogenic effects of TDCPP. Our multi-dimensional profiling approach for organophosphate chemicals can uniquely correlate phenotypic changes with transcriptomic perturbations.

Kidney-Targeted Cytosolic Delivery of siRNA Using a Small-Sized Mirror DNA Tetrahedron for Enhanced Potency.

A proper intracellular delivery method with target tissue specificity is critical to utilize the full potential of therapeutic molecules including siRNAs while minimizing their side effects. Herein, we prepare four small-sized DNA tetrahedrons (sTds) by self-assembly of different sugar backbone-modified oligonucleotides and screened them to develop a platform for kidney-targeted cytosolic delivery of siRNA. An in vivo biodistribution study revealed the kidney-specific accumulation of mirror DNA tetrahedron (L-sTd). Low opsonization of L-sTd in serum appeared to avoid liver clearance and keep its size small enough to be filtered through the glomerular basement membrane (GBM). After GBM filtration, L-sTd could be delivered into tubular cells by endocytosis. The kidney preference and the tubular cell uptake property of the mirror DNA nanostructure could be successfully harnessed for kidney-targeted intracellular delivery of p53 siRNA to treat acute kidney injury (AKI) in mice. Therefore, L-sTd could be a promising platform for kidney-targeted cytosolic delivery of siRNA to treat renal diseases.