Previous studies suggested that microbial communities can harbor keystone species whose removal can cause a dramatic shift in microbiome structure and functioning. Yet, an efficient method to systematically identify keystone species in microbial communities is still lacking. Here, we propose a Data-driven Keystone species Identification (DKI) framework based on deep learning to resolve this challenge. Our key idea is to implicitly learn the assembly rules of microbial communities from a particular habitat by training a deep-learning model using microbiome samples collected from this habitat. The well-trained deep-learning model enables us to quantify the community-specific keystoneness of each species in any microbiome sample from this habitat by conducting a thought experiment on species removal. We systematically validated this DKI framework using synthetic data and applied DKI to analyze real data. We found that those taxa with high median keystoneness across different communities display strong community specificity. The presented DKI framework demonstrates the power of machine learning in tackling a fundamental problem in community ecology, paving the way for the data-driven management of complex microbial communities.
Association of probable post-traumatic stress disorder with dietary pattern and gut microbiome in a cohort of women. Nature Mental Health (Published: October 19, 2023). The paper can be downloaded here.
Post-traumatic stress disorder (PTSD) is a psychiatric condition that may occur in people who have experienced or witnessed traumatic events. The microbiota–gut–brain axis has been suggested to play an important role in mental health. Here we analysed information on trauma exposure and PTSD symptoms with gut microbiome data and dietary information of 191 individuals enrolled in a substudy of an ongoing longitudinal cohort of women. We demonstrated that higher PTSD symptom levels were associated with less adherence to the Mediterranean diet pattern, and this association was also linked to specific PTSD putative protective species such as Eubacterium eligens. Moreover, the microbial pathways involved in the biosynthesis of pantothenate and coenzyme A were identified as PTSD putative protective, and these pathways were mainly contributed by PTSD putative protective species such as Akkermansia muciniphila. These findings have the potential to inform dietary- or microbiome-based interventions for PTSD prevention or amelioration.
We are so pleased to share that our DoD grant has finally been awarded! We will evaluate the PTSD-gut microbiome relationship and develop synbiotics to prevent/ameliorate PTSD. This work is supported by the Office of the Assistant Secretary of Defense for Health Affairs, through the Traumatic Brain Injury and Psychological Health Research Program (Focused Program Award), endorsed by the Department of Defense. In this project, we will collaborate with Drs. Karestan Koenen and Laura Kubzansky from Harvard T.H. Chan School of Public Health, Drs. Jakob Hartmann and Kerry Ressler from McLean Hospital, and Dr. Xuesong He from Forsyth Institute.