Fecal microbiota transplantation (FMT), the transfer of gut microbes from a healthy donor to a patient, has shown promise for ulcerative colitis (UC), but results are inconsistent. While multiple trials report benefit, efforts to pick “good” donors or replace FMT with defined bacterial cocktails have not produced reliable efficacy. A major reason is that we still do not know which donor microbes truly drive long-lasting improvement.
Most studies measure engraftment as whether donor strains are detectable and how abundant they become. Our recent long-read sequencing-based strain tracking work suggests an important missing dimension: after engrafting, bacteria can undergo genomic changes as they adapt to a new host. These changes can include “phase variation,” where genes can switch on or off, potentially altering what a microbe does in the gut even if it remains present.
We will develop an Engraftment Quality Score (EQS) that captures both durable colonization (EQS-D) and genomic stability (EQS-S). Using samples from the FOCUS UC FMT trial and large existing UC datasets, we will identify high-quality strains and functional gene modules, validate them at scale and build explainable machine-learning models to predict response and prioritize candidate strains for next-generation microbiome therapeutics.
