Jeffrey R. Adrion

My Research

I am a postdoctoral associate in the Kern Lab at the University of Oregon. My research lies at the intersection of population genomics, bioinformatics, and computational biology, where I develop machine learning tools that enable biologists to better study evolution.

My work builds on recent breakthroughs in machine leaning technology, in particular the development of deep artificial neural networks, and applies these technologies to pertinent questions in population genomics. Deep learning has recently shown remarkable performance gains in computer vision, speech recognition, natural language processing, and data preprocessing. My research seeks to extend these advancements to topics such as inferring recombination and mutation rates, revealing demographic histories, and characterizing the genomic responses to natural selection. The application of machine learning, and especially of deep learning, to questions in population genomics is ripe with opportunity. The software tools I develop support a simple foundation on which the population genomics community might begin this exploration.

My Software and Collaborations

Image

ReLERNN is a deep learning software pipeline that uses recurrent neural networks to infer the genome-wide landscape of recombination rates directly from raw polymorphism data

Image

TEFLoN is a software pipeline that uses paired-end Illumina sequence data to discover novel transposable element insertions and estimate their allele frequencies

Image

stdpopsim is a community-maintained library of standardized population genetic simulation models spanning diverse taxa

Image

simpoolTE is a program that simulates the insertion and deletion of transposable elements in a pooled population of chromosomes

Image

About Me

I completed my PhD at Indiana University, where I was advised by Matt Hahn and Kristi Montooth. I was broadly trained in the areas of evolutionary biology, population genetics, bioinformatics, and computer science. My dissertation research spanned a diversity of topics, such as characterizing the role of compensatory evolution in shaping mitochondrial-nuclear gene complexes and identifying how spatially varying selection contributed to the genome-wide distribution of transposable elements in Drosophila. At IU I completed formal coursework in bioinformatics and machine learning, and my interest in these areas fueled my desire to develop computational tools that assist researchers in the fields of evolutionary biology and population genomics.

I currently live in Eugene, Oregon, with my wife, Sarah, and our dog, Sandy. I love the outdoors and backcountry exploring. My resume and contact information can be found here .