ARLINGTON, VA, September 3, 2025 – March of Dimes, a national leader in maternal and infant health research, announced today that groundbreaking pregnancy data from its Prematurity Research Center (PRC) network has been chosen for a prestigious US National Science Foundation (NSF) artificial intelligence (AI) pilot program.
Of the ten ‘community’ datasets competitively chosen for the National Artificial Intelligence Research Resource (NAIRR) Pilot—a public AI database aimed at advancing the nation’s AI workforce—March of Dimes was the only organization representing pregnancy or healthcare. This recognition highlights the caliber of the organization’s research and its leadership in harnessing AI to help understand, predict, and prevent preterm birth.
“To be nationally recognized for the quality of our data is incredibly humbling,” said Dr. Emre Seli, Chief Scientific Advisor at March of Dimes. “What started as one small study in a field few imagined would be ideal for AI has grown into a resource that inspires students, researchers, doctors, and computational experts around the world—all working toward breakthroughs to improve health outcomes for moms and babies.”
The selected dataset contains more than 3,500 vaginal swab samples from over 1,200 pregnant individuals, one of the largest and most geographically diverse collections of its kind. It's housed in the March of Dimes Database for Preterm Birth Research, a repository of every piece of molecular data generated by March of Dimes PRCs since 2015. Today, that database spans 79 studies and more than 46,000 experimental samples from nearly 31,000 participants. It's the only public multiomic resource for preterm birth research in the world, bringing together multiple types of biological data including DNA, RNA, immune system markers, and the microbiome. The database itself is managed by scientists at the March of Dimes PRC at the University of California, San Francisco (UCSF).
A key strength of this vaginal microbiome dataset is that it’s “AI-ready", expertly standardized using a tool called MaLiAmPi, developed by UCSF PRC collaborator Dr. Jonathan Golob. By harmonizing data from multiple microbiome studies and making it publicly available on the database, the team opened the door for global AI and health science communities to advance understanding of vaginal microbiome-driven preterm birth, one of the leading causes of spontaneous early labor.
“Our vaginal microbiome dataset is sizable, feature-rich, and well-documented, making it ideal for developing models, teaching AI, and exploring microbial patterns,” said Dr. Tomiko Oskotsky, who co-directs the database at the UCSF PRC. “It’s a rare resource that helps scientists test ideas and tools on real, complex biological data to further our understanding of preterm birth.”
The dataset has already attracted national attention. In 2024, it took center stage in a Cell Reports Medicine paper detailing an open science competition called a DREAM Challenge, where computational scientists from around the world built predictive models for preterm birth. As part of this effort, March of Dimes researchers also developed VMAP (Vaginal Microbiome Atlas in Pregnancy), a visualization tool that opens the data to non-technical users. Now, inclusion in the NAIRR Pilot dramatically expands the data’s reach, enabling collaboration across disciplines to fuel discoveries for moms and babies.
“The incorporation of the March of Dimes microbiome dataset in the NAIRR pilot allows us the opportunity to make an impact beyond the preterm birth research community,” said Dr. Marina Sirota, the UCSF PRC’s lead investigator and database co-director. “This will allow researchers in many other fields to learn more about our data sharing efforts and leverage their AI expertise to use this data for improving diagnostic and therapeutic strategies for preterm birth.”
Launched in 2024, the NAIRR Pilot brings together industry leaders like Microsoft, Google, NVIDIA, and Anthropic with US government agencies including the Department of Defense, the National Oceanic and Atmospheric Administration, and the Department of Energy. It serves as a proof-of-concept for a full-scale NAIRR aimed at strengthening US leadership in AI innovation.
About March of Dimes
March of Dimes leads the fight for the health of all moms and babies. We support research, education, and advocacy and provide programs and services so that every family can get the best possible start. Since 1938, we’ve built a successful legacy to support every pregnant person and every family. Visit marchofdimes.org or nacersano.org for more information.
Contact: Sarah Smith
Email: [email protected]