RUI: enabling phylogenomic research on microbial eukaryotes with eukphylo

The National Science Foundation Division of Biological Infrastructure awarded Smith College $736,602 on January 15, 2027, to expand and refine EukPhylo, an open-source bioinformatic platform for phylogenomic analysis of microbial eukaryotes.

The project develops a user-friendly computational toolkit that integrates best practices from software engineering and incorporates artificial intelligence and machine learning to enable robust, reproducible analysis of genome-scale data from diverse microeukaryotic species including flagellates and amoebae. The platform addresses a gap in existing bioinformatic pipelines, which focus primarily on bacteria or animals and typically require substantial technical expertise, by lowering barriers to entry for scientists with variable bioinformatic skills. EukPhylo will generate standardized results suitable for hypothesis testing on ecologically important and disease-causing microbial species.

Work runs through December 31, 2029, and is performed in Northampton, Massachusetts. The award is funded under the Biological Sciences program (CFDA 47.074).

The project includes workforce development through training of undergraduate and graduate students and a postdoctoral fellow, and will produce written and video tutorials as well as public-facing training sessions to support adoption of the toolkit across the scientific community.

Abstract

The vast majority of species on earth are microbial, including most eukaryotes (i.e. species whose cells have nuclei, like humans). Yet, few tools exist to study diverse microeukaryotes (e.g., flagellates, amoebae), largely because most bioinformatic pipelines focus only on data from bacteria or animals. Many existing platforms also require considerable technical expertise, preventing access for many scientists, and as a result, science in the USA does not efficiently leverage data from microbial species. The work proposed addresses this problem by developing a comprehensive bioinformatic platform, eukphylogenomics (eukphylo), for the analysis of genome-scale data from microbial species. The team will expand and refine eukphylo, an open-source pipeline created at smith college, via the integration of best practices from software engineering. Eukphylo will generate robust, reproducible, and standardized results while also enabling community members with variable bioinformatic skills to analyze diverse datatypes. The proposal also intersects with the NSF priority to advance artificial intelligence (AI): both AI and machine learning will be integrated in the development and deployment of the toolkit, improving reliability and reproducibility. An additional outcome will be the growth of the workforce with expertise in bioinformatics/biotechnology, achieved by training undergraduate and graduate students and a postdoctoral fellow. Other products will include tutorials (written and video) as well as public-facing training sessions. Combined, these efforts will produce a user-friendly toolkit that will transform studies of microscopic species, allowing hypothesis testing on ecologically important groups as well as on the many microbial species that cause diseases in humans. Through the expansion and democratization of the eukphylo cyberinfrastructure, the proposed work will create reproducible and widely accessible bioinformatics workflows that facilitate the curation of data and testing of hypotheses involving eukaryotic microorganisms. The resulting infrastructure will be open-source and fully containerized for easy installation and will come with documentation in the form of written (e.g., publications plus github wiki) and video trainings. Eukphylo will be tested through collaboration with labs throughout the US and beyond while building educational/training modules and tutorials for an expanded range of supported analyses. The work addresses three aims, the first of which is to transform eukphylo into an easy-to-use pipeline for the analysis of eukaryotic gene families, leveraging artificial intelligence and machine learning for the reproducible detection of both contaminants and gene transfers. The second aim is to expand the functionality of eukphylo by adding tools to support additional input data types (e.g., metagenomic and metatranscriptomic) and incorporating new analysis modules (e.g., population-scale analyses, detecting signatures of natural selection). The third aim is to reduce computational barriers for researchers working across disciplines through workshops and the generation of educational materials. In support of all aims, the project will train multiple undergraduate and graduate students, and at least one postdoctoral fellow. Additional outcomes will include educational modules/tutorials for training, which will be aimed at a diversity of levels. Hence the project will transform analyses of the wealth of genome-scale data from microeukaryotic species. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the foundation's intellectual merit and broader impacts review criteria.- subawards are not planned for this award.