Figure 1 in Van Eijnatten et al. (2025): "Consecutive steps of the SpeSpeNet workflow. (1) The user uploads the count table, taxonomy table, and environmental metadata table. (2) The user can choose to aggregate the counts at the genus rank. (3) The user can choose to use all samples or a selection. (4) Next, taxa are filtered on zero counts and relative abundance with user-specified cutoffs. (5) By default, zero counts are substituted with random pseudo-counts below the detection limit to prevent spurious correlations. (6) The data are normalized using TSS or CLR transformation. (7) Calculation of correlation matrix (Spearman, SparCC, Pearson, or Kendall), clusters, and environmental overlay. (8) Network is visualized."
New publication: SpeSpeNet: an interactive and user-friendly tool to create and explore microbial correlation networks, by @jwubs.bsky.social and others. #microbiomedata #correlationnetworks
doi.org/10.1093/isme...