In this role, I helped to define the vision for the use of large language models, ontology, and knowledge graphs to better help life scientists using the Rosalind Research application. In addition, I played an important part in the product development process, leading the transition from a scrum-oriented management style to the formation of cross-functional product-focused teams.

Responsible for both management and hands-on contributions to multiple aspects of the Rosalind Research platform. Projects included the design and implementation of a large foundational biomedical knowledge graph, modeling and prototyping a graph of gene expression experiments and results, and designing new user experiences powered by generative AI and the Rosalind knowledge graph.

My primary role at Data4cure was to enhance the data-driven "CURIE" Knowledge Graph that sits at the center of their 'Biomedical Intelligence Cloud' platform. I worked directly on leveraging biomedical ontologies to help consolidate and semantically organize the knowledge graph, provided technical guidance on various knowledge import projects, and prototyped the use of knowledge graph technologies such as Amazon Neptune.

Helping the Gene Ontology expand the Semantic Web for the life sciences with Causal Activity Models. OWL, RDF, SPARQL, BlazeGraph, Web applications, bioinformatics, Reactome, BioPAX, pathway databases and analysis. More at http://noctua.geneontology.org

Research scientist focused on organizing biomedical knowledge through crowdsourcing, machine learning and citizen science.

Crowdsourcing for bioinformatics research

I studied and developed a variety of crowd/community oriented projects for capturing biomedical knowledge. Efforts of note include the Gene Wiki (done in the context of Wikipedia) and several explorations into games with a scientific purpose.

In this role I worked to enhance and mine the "Gene Wiki"- the collection of articles about genes on Wikipedia. I primarily focused on developing a text-mining pipeline for extracting gene ontology and disease ontology-based annotations from the text of Gene Wiki articles.

trueIDapps develops applications for the Apple iPhone and iPod touch. My role within the company was to design and develop websites and web services that supported the applications. This work was conducted with the Google App Engine for Java.

Freeform Information attempted to bring high quality competitive intelligence data into the hands of biotechnology industry analysts. The distinctive feature of this company was the idea of building a knowledge-based business on top of Freebase - the world's largest open-access database.

My PhD research focused on the basic premise of leveraging the social web (or people, their knowledge, their reasoning) to break through the bottlenecks of build a semantic web (richly structured, web-oriented, data and knowledge) for the life sciences. Paraphrasing the abstract...The rapid expansion of bioinformatics resources on the Web presents significant challenges for biomedical researchers in finding and integrating vital data. The semantic Web initiative offers a solution through standards for sharing distributed information, anchored by ontologies that structure metadata to make biological datasets discoverable and interoperable.This dissertation develops novel, crowdsourcing-driven strategies to amass, evaluate, and apply ontologies and ontology-based annotations in bioinformatics. It introduces designs for systems that engage third-party contributors to collaboratively build collective metadata resources. By incentivizing participation and providing structured guidance, these systems create scalable, high-quality ontologies that evolve with the field. Furthermore, the research introduces fully automated protocols for assessing and comparing ontologies, ensuring their effectiveness and relevance as biological data continues to proliferate.The practical applications of this research showcase how distributed data sources can be dynamically integrated for advanced visualization and analysis, leading to more efficient discovery processes. Ultimately, this work provides a framework for building a semantic Web tailored to life sciences, empowering researchers to collaboratively create, share, and leverage biological knowledge at scale.

Database design and implementation. Web development. Analysis of amino acid sequence data and microarray experiments via machine learning. Worked with genetic information related to HIV and Coccidioides immitis.

Developed Knowledge Bases for the DARPA funded High Performance Knowledge Bases Project (HPKB). Worked closely with Barbara Starr at SAIC and Vinay Chaudhri at the Stanford Research Institute (SRI)