NRMD-based beta diversity analysis

To quantify the similarity/dissimilarity between the different disease cohorts, we used the normalized difference between cases and controls in each disease cohort (NRMD). Since not all ASVs are present in all disease cohorts, for each pair of disease cohorts, we calculated a modified Bray-Curtis distance using only the ASVs present in both disease cohorts. The distance was calculated as follows:

where i and j denote two disease cohorts, and D(i, j) is the modified Bray-Curtis distance between these two disease cohorts.

This modified Bray-Curtis distance is similar to the classic Bray-Curtis distance, but calculated only on ASVs present in both samples. We opted for this modified Bray-Curtis metric to reduce the effect of the different disease cohorts (i.e., different populations/extraction protocols etc.) that can lead to the lack of observations of some ASVs in specific cohorts, therefore prohibiting the determination of the NRMD for the ASV in the specific cohort.

A distance matrix for all disease cohort pairs was calculated using this modified Bray-Curtis metric. This matrix was then used for the generation of a PCoA depicting disease cohort similarity using qiime2.