Author Correction: Quantifying bias introduced by sample collection in relative and absolute microbiome measurements

Subjects

In the original published version of this article, there was a 10-fold error in the 16S rRNA gene standard curves. This was likely caused by a 1:10 serial dilution error of the standard plasmids. The error was determined from comparison of 16S rRNA gene standard curves to subsequent studies of prokaryotic concentration in our group. This discrepancy was found by Boryana Garcia Doyle, who subsequently assisted with our correction of the paper and has now been added as an author. To correct the error, we re-ran all samples through the 16S rRNA qPCR workflow with modifications: 1) For each 16S rRNA standard plasmid, serial dilutions of the plasmids were independently generated in triplicate, totaling six standard curves in each qPCR run, to confirm the standard curve. 2) Samples were run in qPCR technical duplicate instead of triplicate, as in the original manuscript, to accommodate additional wells required for the added replicates of the standard curves. The reported measurements have been updated and conclusions of the paper remain largely unchanged. We updated the conclusion of prokaryotic cell count to 1.86 10¹³and this result is in line with prior published estimates. Additionally, we found a significant difference in total prokaryotic concentration between unpreserved, immediately frozen samples and immediately frozen, Zymo-preserved samples. Figures 4, 5Supplementary Figs. 68and Supplementary Data 13 and Supplementary Table 4 have been updated to reflect these revisions; for comparison, original versions and a list of text edits are available in the online version of this amendment.

Supplementary information is available in the online version of this amendment.