A5010505751- Computational Drug Discovery Methods
- Machine Learning in Materials Science
- Biomedical and Engineering Education
- Parkinson's Disease Mechanisms and Treatments
- Metabolomics and Mass Spectrometry Studies
- Neurological diseases and metabolism
- Genetics, Bioinformatics, and Biomedical Research
- COVID-19 Clinical Research Studies
- SARS-CoV-2 and COVID-19 Research
- Pharmacovigilance and Adverse Drug Reactions
- Hip and Femur Fractures
- Gut microbiota and health
- Drug-Induced Adverse Reactions
- Teaching and Learning Programming
- Microbial Natural Products and Biosynthesis
- Genomics and Rare Diseases
- Health Systems, Economic Evaluations, Quality of Life
- Biosimilars and Bioanalytical Methods
- Microbial Metabolic Engineering and Bioproduction
- Scientific Computing and Data Management
- Asymmetric Hydrogenation and Catalysis
- Pharmaceutical studies and practices
- Software Engineering Techniques and Practices
- Pharmaceutical Economics and Policy
- Molecular spectroscopy and chirality
Stanford University
2015-2023
University of California, San Francisco
2020
Icahn School of Medicine at Mount Sinai
2020
Phoenix Children's Hospital
2020
Engaging, hands-on design experiences are key for formal and informal Science, Technology, Engineering, Mathematics (STEM) education. Robotic video game challenges have been particularly effective in stimulating student interest, but equivalent the life sciences not as developed. Here we present concept of a "biotic project" to motivate learning at interface device engineering (as part cornerstone bioengineering devices course). We provide all course material also efforts adapting project's...
The most rapid path to discovering treatment options for the novel coronavirus SARS-CoV-2 is find existing medications that are active against virus. We have focused on identifying repurposing candidates transmembrane serine protease family member II (TMPRSS2), which critical entry of coronaviruses into cells. Using known 3D structures close homologs, we created seven homology models. also identified a set inhibitor drugs, generated several conformations each, and docked them our used three...
The most rapid path to discovering treatment options for the novel coronavirus SARS-CoV-2 is find existing medications that are active against virus. We have focused on identifying repurposing candidates transmembrane serine protease family member II (TMPRSS2), which critical entry of coronaviruses into cells. Using known 3D structures close homologs, we created seven homology models. also identified a set inhibitor drugs, generated several conformations each, and docked them our used three...
Dysregulated iron or Ca
<h3>Importance</h3> Medical device companies submit premarket approval (PMA) statements to the US Food and Drug Administration (FDA) for of highest-risk class devices. Devices indicated pediatric population that use PMA pathway have not been well characterized or analyzed. <h3>Objective</h3> To identify characterize high-risk devices with age indications derived from statements. <h3>Design, Setting, Participants</h3> In this cross-sectional study statements, those containing...
Linear models offer a robust, flexible, and computationally efficient set of tools for modeling quantitative structure-activity relationships (QSARs) but have been eclipsed in performance by nonlinear methods. Support vector machines (SVMs) neural networks are currently among the most popular accurate QSAR methods because they learn new representations data that greatly improve modelability. In this work, we use shallow representation learning to accuracy L1 regularized logistic regression...
ABSTRACT Gene functional enrichment is a mainstay of genomics, but it relies on manually curated databases gene functions that are incomplete and unaware the biological context. Here we present an alternative machine learning approach, Deep Functional Synthesis (DeepSyn), which moves beyond function to dynamically infer set from its associated network literature data, conditioned disease drug context current experiment. Using knowledge graph with 3,048,803 associations between genes,...
Abstract Massively accumulated pharmacogenomics, chemogenomics, and side effect datasets offer an unprecedented opportunity for drug response prediction, target identification prediction. Existing computational approaches limit their scope to only one of these three tasks, inevitably overlooking the rich connection among them. Here, we propose DrugOrchestra, a deep multi-task learning framework that jointly predicts response, targets effects. DrugOrchestra leverages pre-trained molecular...
Studying analog series to find structural transformations that enhance the activity and ADME properties of lead compounds is an important part drug development. Matched molecular pair (MMP) search a powerful tool for analysis imitates researchers' ability select pairs differ only by small well-defined transformations. Abstraction challenge existing MMP algorithms, which can result in omission relevant, inexact MMPs, inclusion irrelevant, contextually dissimilar MMPs. In this work, we present...
"Knowledge graphs" (KGs) have become a common approach for representing biomedical knowledge. In KG, multiple data sets can be linked together as graph representation, with nodes entities, such "chemical substance" or "genes," and edges predicates, "causes" "treats." Reasoning inference algorithms then applied to the KG used generate new We developed three KG-based question-answering systems part of Biomedical Data Translator program. These are typically tested evaluated using traditional...
Summary Dysregulated iron or Ca 2+ homeostasis has been reported in Parkinson’s disease (PD) models. Here we discover a connection between these two metals at the mitochondria. Elevation of levels causes inward mitochondrial overflow, through an interaction Fe with Mitochondrial Calcium Uniporter. In PD neurons, accumulation-triggered influx across surface leads to spatially confined elevation outer membrane, which is subsequently sensed by Miro1, -binding protein. A Miro1 blood test...
Randomized control trials (RCTs) are the gold standard for clinical to evaluate effectiveness and safety of interventions. Conducting RCTs, however, is not always feasible, particularly rare conditions. Real-world data (RWD), including information from Electronic Health Records (EHR), utilized similar analyses in retrospective data. The United States Food Drug Administration increasingly receptive accepting such as evidence long studies designed conducted robustly. EHR work suffers a...
Clinical trials represent a significant risk in the commercialization of surgical technologies. There is incentive for companies to mitigate their regulatory by targeting 510K over Premarket Approval (PMA) pathways order limit scope, complexity and cost clinical trials. As such, not all will publish data scientific literature.We set out investigate relationship between publication device impact it has on company valuation. We hypothesize that publishing literature correlates with success as...