Liat Shenhav, PhD

We are a computational biology research group focused on developing mathematical models and artificial intelligence (AI) algorithms to enhance the health of women and children, particularly in fertility, pregnancy, and lactation. Our goal is to improve maternal and child health outcomes through rigorous, data-driven insights.

Our bespoke computational methods—combining tensor factorization, time-series analysis, and machine learning, alongside principles of community ecology and complexity theory—are designed to uncover hidden dynamics that distinguish normal from pathological conditions. These algorithms uncover interpretable relationships between multiple omic layers and health-related phenotypes in women and children.

Our research centers on three interconnected areas: the human microbiome, human milk, and the dynamics of pregnancy.

The Human Microbiome: Unlike our DNA, our microbiome, the collection of microbes living in and on our body, is dynamic and subject to frequent changes. Our research explores how the microbiome evolves over time and how these changes can cause or indicate health and disease. To this end, we develop computational methods that serve the broader microbiome community and apply them to elucidate the microbiome's role in key life stages of fertility, pregnancy, and lactation.

Human Milk: Remarkably, we know more about what’s in a strawberry than what’s in human milk. To address this gap, we study the intricate composition and dynamics of human milk across diverse populations worldwide. By applying a systems biology approach that integrates data science with human milk research, we aim to comprehensively understand how the components and dynamics of human milk drive early-life development, including the modulation of the microbiome, and shape long-term health outcomes.

The Dynamics of Pregnancy:  We develop innovative AI approaches to model healthy pregnancy progression and detect deviations from optimal trajectories. Using the eye as a window into future vascular health, we focus on the early detection of hypertensive disorders of pregnancy, such as preeclampsia. By combining high-resolution retinal imaging with topological data analysis and deep learning, our approach enables early risk stratification, paving the way for advanced strategies in preventative care and diagnostics.

Our goal is to deepen the understanding of complex biological systems related to fertility, pregnancy, and lactation, with the ultimate aim of improving maternal and child health outcomes. To achieve this, we integrate extensive data collection with theoretical and data-driven approaches, working closely with experts in obstetrics, pediatrics, ophthalmology, microbiome research, and human milk science. This multidisciplinary approach enables us to discover key mechanisms and identify interpretable biomarkers that drive these dynamic systems.