Thanks for your guidance. After using dada2 denoising, I got denoising stat .qza, table seq.qza and rep seq qaza. It took almost 8 hours. Then I correlate this with sample metada.tsv. I got table.qzv and rep-seq.qzv. When I view table.qzv via the link QIIME 2 View . I have become confused. I have three replicates per each sample but in this table these replicates come in separate sample IDs and their position regarding count is also not the same. I mean there is variations in the sequence count. I have attached the table. qzv (png) and above mentioned replicates manifest file with this message. Kindly view and guide.
Hi @Aqleem12,
I moved this inquiry to its own topic as it was no longer related to the original question. We ask that users start new threads for each new topic rather than tagging on to previous/unrelated ones. Thanks!
I’m not sure I understand your highlighted schematics but I’ll try and explain both those figures.
D01 (forward) and D01 (reverse) from your manifest file are merged at the end of DADA2 so they will become just D01.
Each sample that gets imported into qiime2 must have a unique sample-id, regardless of whether they are replicates of each other or not. The steps you’ve taken so far are agnostic of your experimental design, therefore the scripts will treat each replicate as an unrelated sample. So even if D01 and D02 are exactly the same and come from the same sample from your experiment’s perspective, at this point they are treated as separate samples, just as they were treated when they were on the sequencing machine. This is also why there is natural variability between those reads in your feature table for those replicates. Because they were sequenced individually on the machine. This is exactly what we expect so there is nothing wrong here.
How you play with your data from this one point on depends on how you have your metadata file set. For example, if D01, D02, and D03 are true technical replicates and not just different samples from the same treatment group, you can merge them using the feature-table group command. In this scenario you’ll have to add a new metadata column that the script can use to merge those replicates. For example a new column called ‘Replicates’ can explain that sample-id D01, D02, D03 are all in fact ‘D’. Then by calling on this script you can choose to merge all samples that are defined as ‘D’.
Hope that helps.
