A5000981301- Genetic Associations and Epidemiology
- Genomics and Rare Diseases
- Pharmacovigilance and Adverse Drug Reactions
- Computational Drug Discovery Methods
- Pharmacogenetics and Drug Metabolism
- Misinformation and Its Impacts
- HIV, Drug Use, Sexual Risk
- Coral and Marine Ecosystems Studies
- Sentiment Analysis and Opinion Mining
- Web Data Mining and Analysis
- Microbial Community Ecology and Physiology
- Pharmaceutical studies and practices
- Epigenetics and DNA Methylation
- Advanced Text Analysis Techniques
- Toxin Mechanisms and Immunotoxins
- Tuberculosis Research and Epidemiology
- Bioinformatics and Genomic Networks
- Semantic Web and Ontologies
- Marine and coastal plant biology
- Cancer Genomics and Diagnostics
- Genetics, Bioinformatics, and Biomedical Research
- Biosimilars and Bioanalytical Methods
- BRCA gene mutations in cancer
- Coastal wetland ecosystem dynamics
- Biomedical Text Mining and Ontologies
Stanford University
2017-2025
Colby College
2018-2019
Pharmacogenetics (PGx) studies the influence of genetic variation on drug response. Clinically actionable associations inform guidelines created by Clinical Implementation Consortium (CPIC), but broad impact entire populations is not well understood. We analyzed PGx allele and phenotype frequencies for 487,409 participants in UK Biobank, largest study to date. For 14 CPIC pharmacogenes known human response, we find that 99.5% individuals may have an atypical response at least 1 drug; average...
Protein-truncating variants can have profound effects on gene function and are critical for clinical genome interpretation generating therapeutic hypotheses, but their relevance to medical phenotypes has not been systematically assessed. Here, we characterize the effect of 18,228 protein-truncating across 135 from UK Biobank find 27 associations between in genes outside major histocompatibility complex. We perform phenome-wide analyses directly measure homozygous carriers, commonly referred...
The introduction of frameshift indels by genome editing has emerged as a powerful technique to study the functions uncharacterized genes in cell lines and model organisms. Such mutations should lead mRNA degradation owing nonsense-mediated decay or production severely truncated proteins. Here, we show that engineered can also skipping "multiple three nucleotides" exons. splicing events result in-frame may encode fully partially functional We characterize segregating nonsense variant...
Large biobanks linking phenotype to genotype have led an explosion of genetic association studies across a wide range phenotypes. Sharing the knowledge generated by these resources with scientific community remains challenge due patient privacy and vast amount data. Here, we present Global Biobank Engine (GBE), web-based tool that enables exploration relationship between in biobank cohorts, such as UK Biobank. GBE supports browsing for results from genome-wide studies, phenome-wide...
Pharmacogenomics (PGx) decision support and return of results is an active area precision medicine. One challenge implementing PGx extracting genomic variants assigning haplotypes in order to apply prescribing recommendations information from the Clinical Pharmacogenetics Implementation Consortium (CPIC), US Food Drug Administration (FDA), Knowledgebase (PharmGKB), etc. Annotation Tool (PharmCAT) (i) extracts specified guidelines a genetic data set derived sequencing or genotyping...
The small molecule Retro-2 prevents ricin toxicity through a poorly-defined mechanism of action (MOA), which involves halting retrograde vesicle transport to the endoplasmic reticulum (ER). CRISPRi genetic interaction analysis revealed activity resembles disruption transmembrane domain recognition complex (TRC) pathway, mediates post-translational ER-targeting and insertion tail-anchored (TA) proteins, including SNAREs required for transport. Cell-based in vitro assays show that blocks...
Adverse drug reactions (ADRs) affect the health of hundreds thousands individuals annually in United States, with associated costs billions dollars. The monitoring and analysis severity ADRs is limited by current qualitative categorical systems classification. Previous efforts have generated quantitative estimates for a subset but were scope because time efforts.The aim this study to increase number which there are while improving quality these estimates.We present semisupervised approach...
Abstract Pharmacogenetics (PGx) studies the influence of genetic variation on drug response. Clinically actionable associations inform guidelines created by Clinical Implementation Consortium (CPIC), but broad impact entire populations is not well-understood. We analyzed PGx allele and phenotype frequencies for 487,409 participants in U.K. Biobank, largest study to date. For fourteen CPIC pharmacogenes known human response, we find that 99.5% individuals may have an atypical response at...
Protein-truncating variants can have profound effects on gene function and are critical for clinical genome interpretation generating therapeutic hypotheses, but their relevance to medical phenotypes has not been systematically assessed. We characterized the effect of 18,228 protein-truncating across 135 from UK Biobank found 27 associations between in genes outside major histocompatibility complex. performed phenome-wide analyses directly measured homozygous carriers, commonly referred as...
Abstract Opioid-involved overdose deaths have risen significantly since 1999 with over 80,000 annually 2021, primarily driven by synthetic opioids, like fentanyl. Responding to the rapidly changing opioid crisis requires reliable and timely information. One possible source of such data is social media platforms billions user-generated posts, a fraction which are about drug use. We therefore assessed utility Reddit for surveillance epidemic, covering prescription, heroin, drugs (as September...
Large biobanks linking phenotype to genotype have led an explosion of genetic association studies across a wide range phenotypes. Sharing the knowledge generated by these resources with scientific community remains challenge due patient privacy and vast amount data. Here we present Global Biobank Engine (GBE), web-based tool that enables exploration relationship between in large biobank cohorts, such as UK Biobank. GBE supports browsing for results from genome-wide studies, phenome-wide...
Abstract Genetics plays a key role in drug response, affecting efficacy and toxicity. Pharmacogenomics aims to understand how genetic variation influences response develop clinical guidelines aid clinicians personalized treatment decisions informed by genetics. Although pharmacogenomics has not been broadly adopted into practice, genetics regardless. Physicians adjust patient care based on observed medication, which may occur as result of variants harbored the patient. Here we seek selection...
A delicate relationship exists between reef-building corals and their photosynthetic endosymbionts. Unfortunately, this can be disrupted, with expelling these algae when temperatures rise even marginally above the average summer maximum. Interestingly, several studies indicate that failure of to regulate symbiont cell divisions at high may underlie disruption; increased proliferation symbionts stress host cells by over-production reactive oxygen species or disrupting flow nutrients. This...
Abstract Population-scale biobanks that combine genetic data and high-dimensional phenotyping for a large number of participants provide an exciting opportunity to perform genome-wide association studies (GWAS) identify variants associated with diverse quantitative traits diseases. A major challenge GWAS in population is ascertaining disease cases from heterogeneous sources such as hospital records, digital questionnaire responses, or interviews. In this study, we use parameters including...
Abstract Adverse drug reactions (ADRs) impact the health of 100,000s individuals annually in United States with associated costs hundreds billions. The monitoring and analysis severity adverse is limited by current qualitative categorical system classifications. Previous efforts have generated quantitative estimates for a subset ADRs, but were scope due to time efforts. We present semi-supervised approach that ADR using lexical network word embeddings label propagation. use this method...
Abstract Social media has been identified as a promising potential source of information for pharmacovigilance. The adoption social data hindered by the massive and noisy nature data. Initial attempts to use have relied on exact text matches drugs interest, therefore suffer from gap between formal drug lexicons informal media. Reddit comment archive represents an ideal corpus bridging this gap. We trained word embedding model, RedMed, facilitate identification retrieval health entities...