Objectives Aim 1: To determine differential dynamic temporal relationships between microbial composition/ functional genes and circulating serum metabolites in KTRs with pain vs no pain. H1: Specific gut microbiota phenotypes (e.g., low abundance of Akkermansia, differential beta diversity indices) will be identified in those with pain vs no pain, and microbiota taxa will be associated with serum metabolite levels (e.g., decreased SCFAs, serotonin, kynurenic acid, indoles; increased neurotoxic quinolinic acid, endotoxin, urea).
Aim 2: To determine the moderation effects of diet and stress on dynamic temporal relationships between microbiome features, serum metabolites, and pain scores among KTRs. H2: Those with pain will report higher stress and consume a low fiber diet (e.g., fiber grams per 1000 kcal), resulting in a shift to a proinflammation microbiome phenotype (e.g., lower alpha diversity, lower abundances of Akkermansia, higher Enterococcus), lower serum levels of SCFAs, and higher levels of neurotoxic metabolites (e.g., quinolinic acid).
Aim 3 (exploratory): To identify host-microbial interactions that are causally linked to PI among KTRs. H3: Integration of longitudinal data from biomarkers associated with PI into clinical-based dynamic machine learning models (e.g., race, age, income, kidney function, diet, stress) will improve their accuracy by \>30% as host and microbial biomarkers can better capture environmental factors associated with PI.
