A5065577251- Computational Drug Discovery Methods
- Cholinesterase and Neurodegenerative Diseases
- Neuroscience and Neuropharmacology Research
- Bioinformatics and Genomic Networks
- Protein Structure and Dynamics
- Chemical Synthesis and Analysis
- Metabolomics and Mass Spectrometry Studies
- Machine Learning in Materials Science
- Recommender Systems and Techniques
- Gene expression and cancer classification
- Receptor Mechanisms and Signaling
- Machine Learning in Bioinformatics
- Synthesis and Characterization of Heterocyclic Compounds
- Synthesis and Reactions of Organic Compounds
- Nicotinic Acetylcholine Receptors Study
- Cell Image Analysis Techniques
- Distributed and Parallel Computing Systems
- Phenothiazines and Benzothiazines Synthesis and Activities
- Chemical synthesis and pharmacological studies
- Synthesis and Catalytic Reactions
- Advanced Graph Neural Networks
- Asymmetric Synthesis and Catalysis
- Advanced Fluorescence Microscopy Techniques
- Cloud Computing and Resource Management
- Synthesis and biological activity
Simulations Plus (United States)
2025
Janssen (United States)
2017-2020
Janssen (Belgium)
2017-2019
Springhouse
2019
Lomonosov Moscow State University
2006-2015
Université de Strasbourg
2013-2014
Institute of Physiologically Active Compounds
2010-2013
Middle East Technical University
2013
Centre National de la Recherche Scientifique
2013
Moscow State University
2008
Chemogenomics data generally refers to the activity of chemical compounds on an array protein targets and represents important source information for building in silico target prediction models. The increasing volume chemogenomics offers exciting opportunities build models based Big Data. Preparing a high quality set is vital step realizing this goal work aims compile such comprehensive dataset. This dataset comprises over 70 million SAR points from publicly available databases (PubChem...
We herewith present a novel approach to predict protein-ligand binding modes from the single two-dimensional structure of ligand. Known X-ray structures were converted into binary bit strings encoding interactions. An artificial neural network was then set up first learn and interaction fingerprints simple ligand descriptors. Specific models constructed for three targets (CDK2, p38-α, HSP90-α) 146 ligands which are available. These able discriminate important features minor Predicted...
Artificial intelligence (AI) is undergoing a revolution thanks to the breakthroughs of machine learning algorithms in computer vision, speech recognition, natural language processing and generative modelling. Recent works on publicly available pharmaceutical data showed that AI methods are highly promising for Drug Target prediction. However, quality public might be different than industry due labs reporting measurements, measurement techniques, fewer samples less diverse specialized assays....
A mini-HTS on 4000 compounds selected using 2D fragment-based similarity and 3D pharmacophoric shape to known selective tau aggregate binders identified N-(6-methylpyridin-2-yl)quinolin-2-amine 10 as a novel potent binder human AD aggregated with modest selectivity versus β-amyloid (Aβ). Initial medicinal chemistry efforts key elements for potency selectivity, well suitable positions radiofluorination, leading first generation of fluoroalkyl-substituted quinoline binding ligands suboptimal...
Bayesian matrix factorization is a method of choice for making predictions large-scale incomplete matrices, due to availability efficient Gibbs sampling schemes and its robustness overfitting. In this paper, we consider large scale matrices with high-dimensional side information. However, the link information standard approaches costs O(F <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> ) time, where F dimensionality features. To overcome...
Transthyretin (TTR) is a key transporter of the thyroid hormone thyroxine, and chemicals that bind to TTR, displacing hormone, can disrupt endocrine system, even at low concentrations. This study evaluates computational modeling strategies developed during Tox24 Challenge, using data set 1512 compounds tested for TTR binding affinity. Individual models from nine top-performing teams were analyzed performance uncertainty regression metrics applicability domains (AD). Consensus by averaging...
We describe SILIRID (Simple Ligand-Receptor Interaction Descriptor), a novel fixed size descriptor characterizing protein-ligand interactions. can be obtained from the binary interaction fingerprints (IFPs) by summing up bits corresponding to identical amino acids. This results in vector of 168 integer numbers product number entries (20 acids and one cofactor) 8 types per acid (hydrophobic, aromatic face face, edge H-bond donated protein, ligand, ionic bond with protein cation anion, metal...
We propose Macau, a powerful and flexible Bayesian factorization method for heterogeneous data. Our model can factorize any set of entities relations that be represented by relational model, including tensors also multiple each entity. Macau incorporate side information, specifically entity relation features, which are crucial predicting sparsely observed relations. scales to millions instances, hundred observations, sparse features with dimensions. To achieve the scale up, we specially...
Background: Accurate prediction of absorption, distribution, metabolism and excretion (ADME) properties can facilitate the identification promising drug candidates. Methodology & Results: The authors present Janssen generic Target Product Profile (gTPP) model, which predicts 18 early ADME properties, employs a graph convolutional neural network algorithm was trained on between 1000-10,000 internal data points per predicted parameter. gTPP demonstrated stronger predictive power than...
Asymmetric syntheses of two GlaxoSmithKline's highly potent phosphodiesterase IV inhibitors CMPI 1 and CMPO 2 have been accomplished from nitroethane simple precursors in 8 7 steps, respectively. The suggested synthetic strategy involves as a key stage the silylation enantiopure six-membered cyclic nitronates. In vitro studies PDE IVB1 inhibition revealed significant difference activity enantiomers.
Real-world scientific applications often encompass end-to-end data processing pipelines composed of a large number interconnected computational tasks various granularity. We introduce HyperLoom, an open source platform for defining and executing such in distributed environments providing Python interface tasks. HyperLoom is self-contained system that does not use external scheduler the actual execution task. have successfully employed chemogenomics used pharmaceutic industry novel drug discovery.
We repurpose a High-Throughput (cell) Imaging (HTI) screen of glucocorticoid receptor assay to predict target protein activity in multiple other seemingly unrelated assays. In two ongoing drug discovery projects, our repurposing approach increased hit rates by 60- 250-fold over that the primary project assays while increasing chemical structure diversity hits. Our results suggest data from available HTI screens are rich source information can be reused empower efforts.