Bioinformatics platform for large scale untargeted metabolomics analysis

The project aims to develop an advanced bioinformatics platform to identify and characterize small molecule metabolites: the ubiquitous compounds that compromise everything from life-saving medications to toxic waste. While these compounds are critically important, we have only chemically identified a small fraction of them. Thus, there is a critical need for robust tools to identify and characterize novel metabolites. This proposal will rebuild our existing software to create a new, next-generation platform with the necessary data mining capabilities to analyze small molecule metabolites at the repository scale. From the discovery of new medications to tracking harmful chemicals in the environment, this software will empower researchers to mine historical data to identify novel metabolites, detect harmful compounds, and map chemical exposures. Key software improvements include cutting-edge algorithms such as MASST for database-wide compound searches, library cleaning for robust metabolite annotations, MassQL for scalable identification of entire chemical classes, and Modifinder for predicting compound structure. These tools represent the next step forward in automating compound identification and bringing untargeted metabolomics into the big data age. For instance, the robust computational structure predictions provided by Modifinder will streamline compound characterization, unearthing new insights within molecular networks. One of the core objectives of this project is overcoming the scalability limitations of academic tools. By creating repository-scale solutions, our proposal will bridge the gap between groundbreaking academic innovation and industrial application. The novel visualizations we propose will also significantly improve the user experience, enhancing ease of use without compromising functionality. These advancements will improve analysis speed and efficiency to enable rapid compound identification at scale. To showcase these cutting-edge tools, the project will apply the next-generation platform to the identification of novel natural products and drug metabolites. Over 10,000 plant and 20,000 human samples will be searched for novel molecules that are structurally related to bioactive compounds and known drugs. This analysis will demonstrate the broad applicability of our platform and highlight the capabilities of each tool. Findings will be disseminated via tutorials, white papers, videos, and collaborative efforts with academic partners to encourage widespread adoption among key stakeholders in the environmental and health sciences. The expected outcomes of this proposal is the commercialization of cutting-edge academic tools for accelerating chemical identification and fostering collaborative research in human health. The platform will enhance the detection and characterization of small molecules, with training resources designed to lower the barrier to entry for researchers, driving innovation and discovery in the field.