My research group designs algorithms, statistical methods, and software solutions for problems motivated by biological data. We are particularly focused on the annotation of biological sequences and the accompanying problem of searching for similar sequences within large-scale biological sequence databases, but our work also addresses transposable elements, soil microbiomes, drug discovery, natural language processing, and animal tracking and behavior classification. Projects in our group range from statistical modeling, to indexing and search algorithms, to low-level software optimization, to FPGAs, to deep neural networks, to web services.
Research in the Wheeler lab revolves around development of statistical models, algorithms, and software for problems (primarily) motivated by biological data sets. This work largely focuses on three broad domains: genomics, drug discovery, and animal tracking/behavior. In the area of genomics, they develop probabilistic models and neural networks designed to improve the sensitivity with which genome sequences are labeled, algorithms to perform these tasks more quickly, and methods to reduce false labeling. In the area of drug discovery, they develop machine learning approaches to rapidly explore the binding potential of billions of candidate drugs. In the area of animal tracking/behavior, they develop complex models that can accurately track multiple interacting animals in recorded video, and automatically classify their behavior.