Constructing a network graph of correlations between microbial taxa and fecal metabolites in the gut of captive marmosets (Callithrix jacchus)

The common marmoset (Callithrix jacchus) has a well-characterized gut microbiome in captivity, but how gut microbiota interact with each other over time is not yet known. To help answer this, we characterized interactions in the gut microbiome of the common marmoset by calculating the Spearman correlation coefficient between 16S genera abundance and targeted metabolomics data collected from 10 marmosets (6 males and 4 females). Samples were collected over two 20-day collection periods, for which there was a 3-week gap between the first and second collection period. After processing collected samples, we visualized significant correlations (defined by q < 0.05 after FDR adjustment) in association network graphs to find genera and metabolites that exhibit a high degree of associations, marking them as highly influential in the microbiome. Clostridiaceae Clostridium sensu stricto 1 was among the highest-degree genera, being significantly associated with a large number of metabolites and other genera. It also had a high relative betweenness centrality and a negative correlation with the highly microbially connected genus Paraprevotella, indicating that it could be part of an “information bottleneck” within the microbe-microbe interaction network. Among the high-degree metabolites were the bile acids and several amino acids including ?-Aminobutyric acid (GABA), Taurine, and Lysine. Metabolites shared the strongest degree of correlation with other metabolites, giving credence to the interpretation of correlation as reflecting true covariation over time.

普通狨猴（Callithrix jacchus）在圈养环境中的肠道菌群已得到充分表征，但其肠道菌群随时间的相互作用模式仍未明确。为解答这一科学问题，本研究通过计算16S菌属丰度与从10只狨猴（6雄4雌）中采集的靶向代谢组学数据间的斯皮尔曼相关系数（Spearman correlation coefficient），解析了普通狨猴肠道菌群的互作关系。样本采集分为两个为期20天的阶段，两次采集间隔为3周。对采集样本完成预处理后，我们通过关联网络图可视化了经错误发现率（False Discovery Rate, FDR）校正后q<0.05的显著关联，以筛选出具有高度关联的菌属与代谢物，并将其标记为肠道菌群中具有高影响力的类群。梭菌科梭菌属狭义群1（Clostridiaceae Clostridium sensu stricto 1）是关联度最高的菌属之一，可与大量代谢物及其他菌属呈现显著关联；其相对介数中心性（betweenness centrality）较高，且与微生物关联度较高的帕拉普雷沃菌属（Paraprevotella）呈负相关，提示其可能参与构成微生物-微生物互作网络中的“信息瓶颈”。高关联度代谢物包括胆汁酸与多种氨基酸，如γ-氨基丁酸（γ-Aminobutyric acid, GABA）、牛磺酸及赖氨酸。代谢物间呈现最强的关联相关性，这为“相关性可反映随时间推移的真实共变异”这一解读提供了实证支持。