Prostate cancer is the second most occurring in men worldwide. To better understand mechanisms of tumorigenesis and possible treatment responses, we developed a mathematical model prostate which considers major signalling pathways known to be deregulated. We personalised this Boolean molecular data reflect heterogeneity specific response perturbations patients. A total 488 samples were used build patient-specific models compared available clinical data. Additionally, eight cell line-specific...

In line with recent advances in neural drug design and sensitivity prediction, we propose a novel architecture for interpretable prediction of anticancer compound using multimodal attention-based convolutional encoder. Our model is based on the three key pillars sensitivity: compounds' structure form SMILES sequence, gene expression profiles tumors prior knowledge intracellular interactions from protein-protein interaction networks. We demonstrate that our multiscale (MCA) encoder...

Proteochemometric (PCM) modelling is a computational method to model the bioactivity of multiple ligands against related protein targets simultaneously.

ABSTRACT Drug exposure is a key contributor to the safety and efficacy of drugs. It can be defined using human pharmacokinetics (PK) parameters that affect blood concentration profile drug, such as steady-state volume distribution (VDss), total body clearance (CL), half-life (t½), fraction unbound in plasma (fu) mean residence time (MRT). In this work, we used molecular structural fingerprints, physicochemical properties, predicted animal PK data features model VDss, CL, t½, fu MRT for 1,283...

Abstract Transcriptional perturbation signatures are valuable data sources for functional genomics. Linking to screenings opens the possibility model cellular phenotypes from expression and identify efficacious drugs. We linked transcriptomics LINCS-L1000 project with cell viability information upon genetic (Achilles project) chemical (CTRP screen) perturbations yielding more than 90 000 signature–viability pairs. An integrated analysis showed that signature is a major factor underlying...

Accurate methods to predict solubility from molecular structure are highly sought after in the chemical sciences. To assess state of art, American Chemical Society organized a "Second Solubility Challenge" 2019, which competitors were invited submit blinded predictions solubilities 132 drug-like molecules. In first part this article, we describe development two models that submitted Blind Challenge 2019 but have not previously been reported. These based on computationally inexpensive...

Abstract Structural cardiotoxicity (SCT) presents a high-impact risk that is poorly tolerated in drug discovery unless significant benefit anticipated. Therefore, we aimed to improve the mechanistic understanding of SCT. First, combined machine learning methods with modified calcium transient assay human-induced pluripotent stem cell-derived cardiomyocytes identify nine parameters could predict Next, applied transcriptomic profiling human cardiac microtissues exposed structural and...

Achieving selectivity for small organic molecules toward biological targets is a main focus of drug discovery but has been proven difficult, example, kinases because the high similarity their ATP binding pockets. To support design more selective inhibitors with fewer side effects or altered target profiles improved efficacy, we developed method combining ligand- and receptor-based information. Conventional QSAR models enable one to study interactions multiple ligands single protein target,...

We present a novel approach for the prediction of anticancer compound sensitivity by means multi-modal attention-based neural networks (PaccMann). In our approach, we integrate three key pillars drug sensitivity, namely, molecular structure compounds, transcriptomic profiles cancer cells as well prior knowledge about interactions among proteins within cells. Our models ingest drug-cell pair consisting SMILES encoding and gene expression profile cell predicts an IC50 value. Gene are encoded...

Proteochemometric models of kinases derived from protein fields and ligand 4-point pharmacophoric fingerprints are predictive visually interpretable.

The potential to predict solvation free energies (SFEs) in any solvent using a machine learning (ML) model based on thermodynamic output, extracted exclusively from 3D-RISM simulations water is investigated. models multiple solvents take into account both the solute and description offer possibility SFEs of with root mean squared errors less than 1 kcal/mol. Validations that involve exclusion fractions or clusters solutes exemplify model's capability novel diverse chemical profiles. In...

Field-based proteochemometric modelling predicts activities and visualizes features, which can support the design of more selective protease inhibitors.

Recent methodological advances in deep learning (DL) architectures have not only improved the performance of predictive models but also enhanced their interpretability potential, thus considerably increasing transparency. In context medicinal chemistry, potential to accurately predict molecular properties, chemically interpret them, would be strongly preferred. Previously, we developed accurate multi-task convolutional neural network (CNN) and graph (GCNN) a set diverse intrinsic metabolic...

Abstract Prostate cancer is the second most occurring in men worldwide. To better understand mechanisms of tumorigenesis and possible treatment responses, we developed a mathematical model prostate which considers major signalling pathways known to be deregulated. We personalised this Boolean molecular data reflect heterogeneity specific response perturbations patients. 488 samples were used build patient-specific models compared available clinical data. Additionally, eight...

The authors have obtained spectral‐spatial EPR images of a phantom significantly larger than those previously obtained. Images homogeneous 4.2 cm in diameter and 6.5 length with equivalent to that used for smaller samples give similar linewidth resolution both population distributions width . Spatial appeared modest degradation. the large provide maps magnetic field partially shimmed magnet.

The potential to predict Solvation Free Energies (SFEs) in any solvent using a machine learning (ML) model based on thermodynamic output, extracted from 3D-RISM simulations water is investigated. models multiple solvents take into account both the solute and description offer possibility SFEs of with root mean squared errors less than 1 kcal/mol. Validations that involve exclusion fractions or clusters solutes exemplify model’s capability novel diverse chemical profiles. In addition being...

The potential to predict Solvation Free Energies (SFEs) in any solvent using a machine learning (ML) model based on thermodynamic output, extracted from 3D-RISM simulations water is investigated. models multiple solvents take into account both the solute and description offer possibility SFEs of with root mean squared errors less than 1 kcal/mol. Validations that involve exclusion fractions or clusters solutes exemplify model’s capability novel diverse chemical profiles. In addition being...

Abstract Transcriptomics perturbation signatures are valuable data sources for functional genomic studies. They can be effectively used to identify mechanism of action new compounds and infer activity different cellular processes. Linking phenotypic studies opens up the possibility model selected phenotypes from gene expression predict drugs interfering with phenotype. At same time, close association transcriptomics changes potentially mask compound specific signatures. By linking...

Network diffusion approaches are frequently used for identifying the relevant disease genes and prioritizing drug sensitivity predictions. Majority of these studies rely on networks representing a single type information. However, using multiplex heterogeneous (networks with multiple interconnected layers) is much more informative helps to understand global topology. We built multi-layered network that incorporates information protein-protein interactions, drug-drug similarities, cell...

Network diffusion approaches are frequently used for identifying the relevant disease genes and prioritizing drug sensitivity predictions. Majority of these studies rely on networks representing a single type information. However, using multiplex heterogeneous (networks with multiple interconnected layers) is much more informative helps to understand global topology. We built multi-layered network that incorporates information protein-protein interactions, drug-drug similarities, cell...