A5078251877- Computational Drug Discovery Methods
- Machine Learning in Materials Science
- Cancer Genomics and Diagnostics
- Pharmacogenetics and Drug Metabolism
- Lymphoma Diagnosis and Treatment
- Metabolomics and Mass Spectrometry Studies
- Genomics and Rare Diseases
- Bioinformatics and Genomic Networks
- Chronic Lymphocytic Leukemia Research
- Advanced Graph Neural Networks
- Protein Structure and Dynamics
- AI in cancer detection
- Complex Network Analysis Techniques
- Biomedical Text Mining and Ontologies
- Genomics and Phylogenetic Studies
- Organ Donation and Transplantation
- Advanced Memory and Neural Computing
- Renal and Vascular Pathologies
- Viral-associated cancers and disorders
- Internet Traffic Analysis and Secure E-voting
- Machine Learning and ELM
- Phenothiazines and Benzothiazines Synthesis and Activities
- Advanced Proteomics Techniques and Applications
- Cutaneous lymphoproliferative disorders research
- Language and cultural evolution
Washington University in St. Louis
2013-2021
James S. McDonnell Foundation
2016-2018
National Human Genome Research Institute
2016
University of Tulsa
2007
Understanding how xenobiotic molecules are metabolized is important because it influences the safety, efficacy, and dose of medicines they can be modified to improve these properties. The cytochrome P450s (CYPs) proteins responsible for metabolizing 90% drugs on market, many computational methods predict which atomic sites a molecule--sites metabolism (SOMs)--are during CYP-mediated metabolism. This study improves prior predicting SOMs by using new descriptors machine learning based neural...
Transplantable kidneys are in very limited supply. Accurate viability assessment prior to transplantation could minimize organ discard. Rapid and accurate evaluation of intra-operative donor kidney biopsies is essential for determining which eligible transplantation. The criterion accepting or rejecting relies heavily on pathologist determination the percent glomeruli (determined from a frozen section) that normal sclerotic. This percentage critical measurement correlates with transplant...
Following automated variant calling, manual review of aligned read sequences is required to identify a high-quality list somatic variants. Despite widespread use in analyzing sequence data, methods standardize have not been described, resulting high inter- and intralab variability.This standard operating procedure (SOP) consists annotate variants with four different calls 19 tags. The indicate reviewer's confidence each the tags commonly observed sequencing patterns artifacts that inform...
Abstract Summary: Cytochrome P450 enzymes (P450s) are metabolic that process the majority of FDA-approved, small-molecule drugs. Understanding how these modify molecule structure is key to development safe, effective XenoSite server an online implementation XenoSite, a recently published computational model for metabolism. predicts which atomic sites molecule—sites metabolism (SOMs)—are modified by P450s. accepts input in common chemical file formats including SDF and SMILES provides tools...
BackgroundPathologist evaluation of donor liver biopsies provides information for accepting or discarding potential livers. Due to the urgent nature decision process, this is regularly performed using frozen sectioning at time biopsy. The percent steatosis in a biopsy correlates with transplant outcome, however there significant inter- and intra-observer variability quantifying steatosis, compounded by section artifact. We hypothesized that deep learning model could identify quantify...
Metabolism of drugs affects their absorption, distribution, efficacy, excretion, and toxicity profiles. is routinely assessed experimentally using recombinant enzymes, human liver microsome, animal models. Unfortunately, these experiments are expensive, time-consuming, often extrapolate poorly to humans because they fail capture the full breadth metabolic reactions observed in vivo. As a result, pathways leading formation toxic metabolites missed during drug development, giving rise costly...
ProteomeScout (https://proteomescout.wustl.edu) is a resource for the study of proteins and their post-translational modifications (PTMs) consisting database PTMs, repository experimental data, an analysis suite PTM experiments, tool visualizing relationships between complex protein annotations. The compendium public coupled with user-uploaded data. provides tools datasets, including summary views subset selection, which can identify within subsets data by testing statistically significant...
Scientists rely on high-throughput screening tools to identify promising small-molecule compounds for the development of biochemical probes and drugs. This study focuses identification promiscuous bioactive compounds, which are that appear active in many experiments against diverse targets but often false-positives may not be easily developed into successful probes. These can exhibit bioactivity due nonspecific, intractable mechanisms action and/or by interference with specific assay...
Machine learning, combined with a proliferation of electronic healthcare records (EHR), has the potential to transform medicine by identifying previously unknown interventions that reduce risk adverse outcomes. To realize this potential, machine learning must leave conceptual `black box' in complex domains overcome several pitfalls, like presence confounding variables. These variables predict outcomes but are not causal, often yielding uninformative models. In work, we envision...
Scaffold network generator (SNG) is an open-source command-line utility that computes the hierarchical of scaffolds define a large set input molecules. networks are useful for visualizing, analysing and understanding chemical data increasingly available through public repositories like PubChem. For example, some groups have used scaffold to identify missed-actives in high-throughput screens small molecules with bioassays. Substantially improving on existing software, SNG robust enough work...
A collection of new approaches to building and training neural networks, collectively referred as deep learning, are attracting attention in theoretical chemistry. Several groups aim replace computationally expensive ab initio quantum mechanics calculations with learned estimators. This raises questions about the representability complex chemical systems networks. Can local-variable models efficiently approximate nonlocal features? Here, we find that convolutional architectures, those only...
Tags or observable features shared by a group of similar agents are effectively used in real and artificial societies to signal intentions can be infer unobservable properties choose appropriate behaviors. Use tags select partners has been shown produce stable cooperation agent populations playing the Prisoner's Dilemma game. Existing tag mechanisms, however, promote only if that requires identical actions from all members. We propose more general tag-based interaction scheme facilitates...
Small-molecule screens are an integral part of drug discovery. Public domain data in PubChem alone represent more than 158 million measurements, 1.2 molecules, and 4300 assays. We conducted a global analysis these data, building network assays connecting the if they shared nonpromiscuous active molecules. This spans both phenotypic target-based screens, recapitulates known biology, identifies new polypharmacology. Phenotypic extremely important for discovery, contributing to discovery large...
Graph algorithms are key tools in many fields of science and technology. Some these depend on propagating information between distant nodes a graph. Recently, there have been number deep learning architectures proposed to learn undirected graphs. However, most aggregate the local neighborhood node, therefore they may not be capable efficiently long-range information. To solve this problem we examine recently architecture, wave, which propagates back forth across an graph waves nonlinear...
Many scientific questions are best approached by sharing data—collected different groups or across large collaborative networks—into a combined analysis. Unfortunately, some of the most interesting and powerful datasets—like health records, genetic data, drug discovery data—cannot be freely shared because they contain sensitive information. In many situations, knowing if private datasets overlap determines it is worthwhile to navigate institutional, ethical, legal barriers that govern access...
Abstract Transplantable kidneys are in very limited supply. Accurate viability assessment prior to transplantation could minimize organ discard. Rapid and accurate evaluation of intra-operative donor kidney biopsies is essential for determining which eligible transplantation. The criteria accepting or rejecting relies heavily on pathologist determination the percent glomeruli (determined from a frozen section) that normal sclerotic. This percentage critical measurement correlates with...
Abstract CIViC is an expert crowdsourced knowledgebase for Clinical Interpretation of Variants in Cancer ( www.civicdb.org ) describing the therapeutic, prognostic, and diagnostic relevance inherited somatic variants all types. committed to open source code, access content, public application programming interfaces (APIs), provenance supporting evidence allow transparent creation current accurate variant interpretations use cancer precision medicine.
In this study, we propose a new, secure method of sharing useful chemical information from small-molecule libraries, without revealing the structures libraries' molecules. Our shares relationship between molecules rather than structural descriptors. This is an important advance because, over past few years, several groups have developed and published new methods analyzing screening data. These include advanced hit-picking protocols, promiscuous active filters, economic optimization...
Pediatric patients are at elevated risk of adverse drug reactions, and there is insufficient information on safety in children. Complicating assessment children, numerous age-dependent changes the absorption, distribution, metabolism, elimination drugs. A key contributor to toxicity ontogeny metabolism enzymes, both abundance type throughout development from fetal period through adulthood. Critically, these affect not only overall clearance drugs but also exposure individual metabolites. In...
In neuroscience, collaboration and data sharing are undermined by concerns over the management of protected health information (PHI) personal identifying (PII) in neuroimage datasets. The HIPAA Privacy Rule mandates measures for preservation subject privacy neuroimaging studies. Unfortunately researcher, is a burdensome task. Wide scale neuroimages challenging three primary reasons: (i) A dearth tools to systematically expunge PHI/PII from sets, (ii) facility tracking patient identities...
Atom- or bond-level chemical properties of interest in medicinal chemistry, such as drug metabolism and electrophilic reactivity, are important to understand predict across arbitrary new molecules. Deep learning can be used map molecular structures their properties, but the data sets for these tasks relatively small, which limit accuracy generalizability. To overcome this limitation, it would preferable model on basis underlying quantum characteristics small However, is difficult learn...
Computing quantum chemical properties of small molecules and polymers can provide insights valuable into physicists, chemists, biologists when designing new materials, catalysts, biological probes, drugs. Deep learning compute accurately in a fraction time required by commonly used methods such as density functional theory. Most current approaches to deep chemistry begin with geometric information from experimentally derived molecular structures or pre-calculated atom coordinates. These have...