Fig. 1

Gut microbiota compositional changes during gestation follow several predictable trends. 16S sequencing of fecal samples from 34 mother sows sampled longitudinally during the 114-day gestation reveals several trends in gut microbiota changes. A A maturity index trained on 60% of the animals and tested on the remaining 40% shows that the amount of time of gestation can be predicted (P = 3.3e−13) with some accuracy (R²=0.27) using gut microbiota composition data. B The 10 taxa that contributed the most to the accuracy of the maturity index are shown in order of importance. C Dirichlet multinomial mixtures (DMM) bin samples into one of 8 clusters, each defined by a unique gut microbiota composition. Days of gestation are on the X-axis, and clusters are labeled on the Y-axis. The total number of samples at each time point is also labeled on the X-axis. The size of the blue circle is proportional to the number of samples contained in each cluster. Transitions made by a substantial percentage of individuals (transition frequency >20%) are indicated by lines connecting blue clusters. D The 10 taxa that contributed the most to the accuracy of the DMM are shown in order of importance. E Heatmap shows the relative abundance of taxa (normalized across each taxa) for each sample. Samples are grouped by DMM cluster to allow for visualization of the taxa that are enriched or reduced on average in each DMM cluster