Malabar project: samples versus otus contingency tables

Malabar project: samples versus otus contingency tables For a short presentation of Malabar project, click here . This dataset makes available a couple of contingency tables with samples as rows and OTUs as columns, or vice-versa, as datasets for further studies in statistical ecology. Main environmental factors are four seasons: summer, autumn, winter, spring four locations: Bouée 13, Comprian, Jaquets, Teychan high tide and low tide pelagos or benthos How the tables have been built We have the fasta files per sample, for two markes: rbcL, 18S-v4. The following steps have been done: All the reads from all samples have been mixed together like in a big bag, one for rbcL, one for 18S-v4, keeping track (through their name) of the sample they come from. For each marker, OTUs have been built (one set of oTUs per "bag") with SWARM, V2. We have a bag of 14,635,946 reads for rbcL, and of 11,826,308 reads for 18S-v4. (such a huge number of reads could be handled with SWARM in a reasonable time). As we know from which sample each read comes, it has been possible to build a contingency table with, say, $n(i,j)$ being the number of reads belonging to OTU $i$ in environmental conditions $j$.