Biology subjects

Ellis, S. B.

Publications and source records attributed to Ellis, S. B..

3 recordsLinked to original sources

bioRxiv2020

Multi-ancestry Genome- and Phenome-wide Association Studies of Diverticular Disease in Electronic Health Records with Natural Language Processing enriched phenotype algorithm

Background and aimsDiverticular disease is among the most prevalent conditions encountered by gastroenterologists, affecting [~]50% of Americans before the age of 60. Our aim was to identify genetic risk variants and clinical phenotypes associated with diverticular disease, utilizing the electronic health record (EHR) with Natural Language Processing (NLP). MethodsWe developed a NLP-enriched phenotype algorithm that incorporated colonoscopy or abdominal imaging reports to accurately identify patients with diverticulosis and diverticulitis from multicenter EHRs. We performed genome-wide association studies (GWAS) of diverticular disease in European, African and multi-ancestry participants, followed by phenome-wide association studies (PheWAS) of the risk variants to identify their potential comorbid/pleiotropic effects in the clinical phenome. For more in-depth investigation of associated clinical phenotypes, we also performed PheWAS with the previously reported 52 GWAS susceptibility variants for diverticular disease. ResultsAncestry-stratified GWAS analyses confirmed the well-established associations between ARHGAP15 loci with diverticular disease in European cohorts, and found similar positive effect sizes in African cohorts but with non-significant p-values. With overall intensified GWAS signals in diverticulitis patients compared to diverticulosis patients, we found substantial genetic correlations between diverticulosis and diverticulitis, up to 0.997 in European ancestry. PheWAS analyses identified associations between the diverticular disease GWAS variants and circulatory system, genitourinary, and neoplastic EHR phenotypes. ConclusionOur multiancestry GWAS-PheWAS study demonstrated an effective use of multidimensional EHR information in disease case/control classification with NLP for more comprehensive and scalable phenotyping, and implementation of an integrative analytical pipeline to facilitate etiological investigation of a disease from a clinical perspective.

bioinformatics↗

bioRxiv2019

Rapid response to the Alpha-1 Adrenergic Agent Phenylephrine in the Perioperative Period is Impacted by Genomics and Ancestry

BackgroundThe emergence of genomic data in biobanks and health systems offers new ways to derive medically important phenotypes, including acute phenotypes that occur during in-patient clinical care. We hypothesized that there is a genetic underpinning to the magnitude of the response to phenylephrine, an 1-adrenergic receptor agonist commonly used to treat hypotension during anesthesia and surgery.\n\nMethodsWe quantified the response to phenylephrine by determining the delta between the minimum blood pressure (BP) within five minutes before and the maximum BP within five minutes after bolus administration. We then performed a genome-wide association study (GWAS) adjusted for genetic ancestry, demographics, and relevant clinical covariates to investigate genetic factors underlying individual differences systolic BP response to phenylephrine ({Delta}SBP), as well as mean arterial pressure ({Delta}MAP) and diastolic BP ({Delta}DBP), for both the entire study cohort as well as for each of 3 ancestry sub-cohorts; European American(EA), African American(AA), and Hispanic American(HA).\n\nResults4,317 patients met inclusion criteria, of which 3,699 were genotyped. Average {Delta}BP values over the entire cohort were {Delta}SBP=17(+-25) mmHg, {Delta}MAP=14(+-18) mmHg, {Delta}DBP=11(+-14) mmHg. The largest difference between populations was observed for {Delta}SBP ({Delta}SBPEA=20(+-24) mmHg; {Delta}SBPHA=16(+-25) mmHg; {Delta}SBPAA=15(+-25) mmHg). The differences remained after adjusting for clinical covariates and ancestry (EA vs. HA: {Delta}SBP, p<0.032;{Delta}MAP, p<0.021;{Delta}DBP,p<0.008);(EA vs. AA:{Delta}SBP,p<5.13x10-5;{Delta}MAP,p<2.1x10-4;{Delta}DBP,p<3.3x10-4). GWAS revealed significant associations between loci and BP response in 5 different genome regions (p<5x10-8) in the entire cohort, and suggestive associations in 2 different regions in EAs (p<6x10-8,p<7x10-8). We observed non-random enrichment in association with SBP drug response in 165 loci previously reported to be associated with systolic blood pressure. Finally, we discovered rare variants, rs188427942 and rs147664194 present at [~]1% in EAs and rs146535276 present at [~]1% in AAs respectively, where patients carrying one copy of these variants show no response to phenylephrine.\n\nConclusionsIt is possible to derive a quantitative phenotype suited for comparative statistics and genome-wide association studies from routinely collected perioperative data. There are population differences in rapid response to phenylephrine, large effect alleles and novel genes affecting pharmaceutical response, and phenylephrine non-responders, with implications for personalized treatment during surgery.

genomics↗

bioRxiv2019

Augmented Intelligence with Natural Language Processing Applied to Electronic Health Records is Useful for Identifying Patients with Non-Alcoholic Fatty Liver Disease at Risk for Disease Progression

ObjectiveElectronic health record (EHR) systems contain structured data and unstructured documentation. Clinical insights can be derived from analyzing both but optimal methods for this have not been studied extensively. We compared various approaches to analyzing EHR data for non-alcoholic fatty liver disease (NAFLD).\n\nMaterials and MethodsWe compared analysis of structured and unstructured EHR data using natural language processing (NLP), free-text search, and diagnostic codes against expert adjudication as the reference standard.\n\nResultsOut of 38,575 patients, we identified 2,281 patients with NAFLD. From the remainder, 10,653 patients with similar data density were selected as a control group. NLP was more sensitive than ICD and text search (NLP 0.93 vs. ICD 0.28 vs. text search 0.81) with higher a F2 score (NLP 0.92 vs. ICD 0.34 vs. text search 0.81). 619 patients had suspected NAFLD documented in radiology notes not acknowledged in other forms of clinical documentation. Of these, 232 (37.5%) were found to have more advanced liver disease after a median of 1,057 days.\n\nDiscussionNLP-based approaches have superior accuracy in identifying NAFLD within the EHR compared to ICD/text search-based approaches. Suspected NAFLD on imaging is often not acknowledged in subsequent clinical documentation. Many such patients are later found to have more advanced liver disease.\n\nConclusionFor identification of NAFLD, NLP performed better than alternative selection modalities and facilitated follow-on analysis of information flow. If accuracy can be proven to persist across clinical domains, NLP can identify patient phenotypes for biomedical research in an accurate and high-throughput manner.

bioinformatics↗