ASD-16S preprocessing OTU tables (complete multiview dataset)

Amplicon sequence data preprocessing was performed using QIIME 2, version 2021.2. Forward sequences were denoised with DADA2, truncated at 240 bp and closed-reference clustered against SILVA database by varying the similarity threshold in { 90%, 94%, 97%, 99% }. Each of the produced OTU tables after clustering was than subjected to feature filtering, that has been performed according to the following criteria:- by <i>frequency</i>, i.e., on the minimum number of individuals for whom a given feature must be present, in { 2, 5, 10, 50 };by <i>samples</i>, i.e., on the minimum percentage of the individuals for whom a given feature must be present, in { 2%, 5%, 10%, 20%, 50% }.<br>Moreover, unfiltered OTU tables are also included. In total, 40 different OTU tables were obtained, considering 4 different similarity thresholds in the clustering phase, multiplied by 10 different filtering configurations, i.e., 4 (filtering by frequency) + 5 (filtering by samples) + 1 (unfiltered).