Pejaver Laboratory

The Pejaver Laboratory

We develop, adapt and apply machine learning methods to relate genetic

variation to molecular function and disease using secondary biomedical data sets

What We Do

Dna icons created by Eucalyp - Flaticon

Genome Interpretation

We integrate functional annotations and predictions to infer the impact of genetic variants individually and jointly over a genome on molecular and organismal phenotypes

Vaccine record icons created by pongsakornRed - Flaticon

Data-driven Phenotyping

We extract and model information from electronic health records (EHRs) to derive quantitative definitions of disease to aid in biomedical discovery and clinical diagnosis

Cluster icons created by Freepik - Flaticon

Applied Machine Learning

We develop and adapt machine learning approaches that specifically address challenges relevant to biomedical data such as noise, bias and limited availability

 

News and Highlights

Multiple positions available

We are actively recruiting postdoctoral fellows and graduate students for a variety of projects. If you are interested in the intersection of biomedical data science, genomic medicine and machine learning and are looking for a supportive training environment, come join us!

COVID-19 prediction DREAM Challenge paper published

Our paper titled “A continuously benchmarked and crowdsourced challenge for rapid development and evaluation of models to predict COVID-19 diagnosis and hospitalization” was just published in JAMA Network Open.

Published: natural language processing for lower back pain

Our paper titled “A comparison of natural language processing methods for the classification of lumbar spine imaging findings related to lower back pain” was just published in Academic Radiology.