By changing variables from false to true in optional_analyses.txt, waterway can rerun specific code blocks and analyses without having to run a whole reanalysis.
Keep in mind that the outputs from most analyses will overwrite any files with the same name in the output folder. To avoid this, move all the files that would be overwritten into a seperate folder.
For more information on specific code and Qiime2 commands, search the “Functions Included” page for the commands listed in the code block.
Setting extended_alpha to true calculates multiple different alpha diversity metrics for the dataset. This will output all QIIME2 files in a folder named “alpha_diversities” in the post-DADA2 outputs folder. The types of alpha diversity calculated include: ace, berger_parker_d, brillouin_d, chao1, chao1_ci, dominance, doubles, enspie, esty_ci, fisher_alpha, gini_index, goods_coverage, heip_e, kempton_taylor_q, lladser_ci, lladser_pe, margalef, mcintosh_d, mcintosh_e, menhinick, observed_otus, osd, pielou_e, robbins, shannon, simpson, simpson_e, singles, and strong. For documentation on the main commands, look here.
If rerun_beta_analysis is set to true, beta diversity comparisons for the groups specified in the variable rerun_group. The groups should be separated by spaces in rerun_group, and must be categorical (not continuous, i.e. not numeric). For example, if the metadata columns “sex”, “diet”, and “coat_color” were to be analysed, the variable should be:
rerun_group=(sex diet coat_color)
This analysis will create a folder called “beta_div_reruns” in the post-DADA2 outputs folder, containing all QIIME2 files describing the analyses, separated by group. The beta diversity is calculated as unweighted and weighted UniFrac.
Similarly, run_beta_continuous can be set to true to calculate beta diversity comparisons for groups specified in the variable continuous_group, again with the groups separated by spaces. However, this comparison if specifically for continuous (numeric) metadata, and therefore is a two-sided Mantel test that compares two distance matrices. This will create the folder “rerun_beta_continuous”, again in the post-DADA2 outputs folder.
ANCOM analyses can be conducted for the dataset by setting run_ancom to true, and by providing the groups of interest to group_to_compare as a space-separated list. The comparisons can be conducted at multiple phylogenetic levels, from level 1 (taxa combined by kingdom) to level 7 (taxa completely separate at the species level) by providing the levels to collapse_taxa_to_level as a space-separated list. All QIIME2 files are stored in a folder named “ancom_outputs” in the post-DADA2 outputs folder. For the QIIME2 documentation, look here.
If PICRUSt2 for QIIME2 is installed in your QIIME2 conda environment, PICRUSt2 can be run on your dataset by setting run_picrust to true. The results will be stored in a folder called “q2-picrust_output” in the post-DADA2 outputs folder.
Inside, the folder “exported_tsv” contains the tabulated output from “pathway_abundance.qzv”, while “exported_tsv_ec” and “exported_tsv_ko” contain the tabulated output from “ec_metagenome.qzv” and “ko_metagenome.qzv” respectively. Core metrics functions are also conducted on the PICRUSt2 output, allowing for 3D PCoA plots calculated on the predicted pathways to be viewed under “pathabun_core_metrics/jaccard_emperor.qzv”.