A5068091427- Computational Drug Discovery Methods
- Analytical Chemistry and Chromatography
- Ultrasound and Cavitation Phenomena
- Scientific Computing and Data Management
- Spectroscopy and Chemometric Analyses
- Face and Expression Recognition
- Hydrogen's biological and therapeutic effects
- Chemical Synthesis and Analysis
- Image Retrieval and Classification Techniques
- Genetics, Bioinformatics, and Biomedical Research
- Medical Image Segmentation Techniques
- Ultrasound and Hyperthermia Applications
- Machine Learning in Materials Science
- Protein Structure and Dynamics
- Metabolomics and Mass Spectrometry Studies
- Biomedical Text Mining and Ontologies
- Cholinesterase and Neurodegenerative Diseases
- Video Surveillance and Tracking Methods
- Distributed and Parallel Computing Systems
- Bioinformatics and Genomic Networks
- Various Chemistry Research Topics
- Spectroscopy and Quantum Chemical Studies
- Web Data Mining and Analysis
- RNA and protein synthesis mechanisms
- Electrochemical Analysis and Applications
Johnson & Johnson (United States)
2007-2021
Janssen (United States)
2020-2021
Janssen (Belgium)
2007-2015
Lomonosov Moscow State University
1992-1998
We present ABCD, an integrated drug discovery informatics platform developed at Johnson & Pharmaceutical Research Development, L.L.C. ABCD is attempt to bridge multiple continents, data systems, and cultures using modern information technology provide scientists with tools that allow them analyze multifactorial SAR make informed, data-driven decisions. The system consists of three major components: (1) a warehouse, which combines from chemical pharmacological transactional databases,...
Multidimensional scaling (MDS) is a collection of statistical techniques that attempt to embed set patterns described by means dissimilarity matrix into low-dimensional display plane in way preserves their original pairwise interrelationships as closely possible. Unfortunately, current MDS algorithms are notoriously slow, and use limited small data sets. In this article, we present family combine nonlinear mapping with neural networks, make possible the very large sets intractable...
We present a novel approach for enhancing the diversity of chemical library rooted on theory wisdom crowds. Our was motivated by desire to tap into collective experience our global medicinal chemistry community and involved four basic steps: (1) Candidate compounds acquisition were screened using various structural property filters in order eliminate clearly nondrug-like matter. (2) The remaining clustered together with in-house collection fingerprint-based clustering algorithm that...
Efficient substructure searching is a key requirement for any chemical information management system. In this paper, we describe the search capabilities of ABCD, an integrated drug discovery informatics platform developed at Johnson & Pharmaceutical Research Development, L.L.C. The solution consists several algorithmic components: 1) pattern mapping algorithm solving subgraph isomorphism problem, 2) indexing scheme that enables very fast searches on large structure files, 3) incorporation...
Protein loops, the flexible short segments connecting two stable secondary structural units in proteins, play a critical role protein structure and function. Constructing chemically sensible conformations of loops that seamlessly bridge gap between anchor points without introducing any steric collisions remains an open challenge. A variety algorithms have been developed to tackle loop closure problem, ranging from inverse kinematics knowledge-based approaches utilize pre-existing fragments...
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...
Producing good low-dimensional representations of high-dimensional data is a common and important task in many mining applications. Two methods that have been particularly useful this regard are multidimensional scaling nonlinear mapping. These attempt to visualize set objects described by means dissimilarity or distance matrix on display plane way preserves the proximities whatever extent possible. Unfortunately, most known algorithms quadratic order, their use has limited relatively small...
Multidimensional scaling (MDS) is a collection of statistical techniques that attempt to embed set patterns described by means dissimilarity matrix into low-dimensional display plane in way preserves their original pairwise interrelationships as closely possible. Unfortunately, current MDS algorithms are notoriously slow, and use limited small data sets. In this article, we present family combine nonlinear mapping with neural networks, make possible the very large sets intractable...
ADVERTISEMENT RETURN TO ISSUEPREVArticleNEXTTemperature Effect on the Sonolysis of Methanol/Water MixturesDmitrii N. Rassokhin, Georgii V. Kovalev, and Lenar T. BugaenkoCite this: J. Am. Chem. Soc. 1995, 117, 1, 344–347Publication Date (Print):January 1995Publication History Published online1 May 2002Published inissue 1 January 1995https://pubs.acs.org/doi/10.1021/ja00106a037https://doi.org/10.1021/ja00106a037research-articleACS PublicationsRequest reuse permissionsArticle...
The utility of chemoinformatics systems depends on the accurate computer representation and efficient manipulation chemical compounds. In such systems, a small molecule is often digitized as large fingerprint vector, where each element indicates presence/absence or number occurrences particular structural feature. Since in theory unique features can be exceedingly large, these vectors are usually folded into much shorter ones using hashing modulo operations, allowing fast "in-memory"...
We present a novel class of topological molecular descriptors, which we call power keys. Power keys are computed by enumerating all possible linear, branch, and cyclic subgraphs up to given size, encoding the connected atoms bonds into two separate components, recording number occurrences each subgraph. have applied these new descriptors for screening stage substructure searching on relational database about 1 million compounds using diverse set reference queries. The can eliminate vast...
We measured the composition of droplets aerosols generated by ultrasound from aqueous solutions containing various concentrations Triton X-100 (an octylphenyl ethoxylate nonionic surfactant) and d-glucose, latter being a reference nonsurfactant solute. To prevent sonochemical decomposition solutes, sonication was performed under atmosphere carbon dioxide. It found that concentration surfactant in ultrasonically aerosol significantly (up to 10 times) higher than solution atomized, while...
A novel approach for selecting an appropriate bin size cell-based diversity assessment is presented. The method measures the sensitivity of index as a function grid resolution, using box-counting algorithm that reminiscent those used in fractal analysis. It shown relative variance score (sum squared cell occupancies) several commonly molecular descriptor sets exhibits bell-shaped distribution, whose exact characteristics depend on distribution data set, number points considered, and...
A general algorithm for the prioritization and selection of plates high-throughput screening is presented. The method uses a simulated annealing to search through space plate combinations one that maximizes some user-defined objective function. robust convergent, permits simultaneous optimization multiple design objectives, including molecular diversity, similarity known actives, predicted activity or binding affinity, many others. It shown arrangement compounds among may have important...