Shallow Shotgun Metagenomic Dataset
We weighed 0.16-g portions of freshly thawed fecal samples into 2-ml bead beating tubes containing Qiagen 0.7-mm dry garnet beads. Next, we added 680 µl of lysis buffer (pre-warmed to 60 ºC) from Qiagen’s QIAamp 96 PowerFecal QIAcube HT Kit to each tube, and bead beat samples for 10 min on a Vortex-Genie 2 fitted with Qiagen’s Vortex Adapter (cat. no 13000-V1-24). All other extraction steps were completed as per Qiagen kit QIAcube HT recommendations (using a single-tube start protocol). Final DNA extracts were eluted in 100 µl of molecular grade water. A total of 2820 samples were extracted using this protocol, as part of a larger study. For this validation, we focus only an 83-sample subset of this larger sample-set. This 83-sample subset represents DNA extracts from technical replicates of fecal samples, which correspond to the 83 fecal samples included in an amplicon dataset (described below).
DNA extracts were quantified using Qubit dsDNA Broad Range Assays. Extracts were sent to the University of Calgary Centre for Health Genomics and Informatics, where shotgun metagenomic libraries were created using iGenomx Riptide preparations (1:1 mixture of low GC and High GC primers), as per manufacturer specifications. Unlike the Nextera XT protocol, the iGenomx Riptide protocol allows for multiplexing of up to 960 samples on a single Illumina sequencing lane. For this pilot study, we sequenced a total of 188 samples using an Illumina NovaSeq (300 cycle SP sequencing kit v1.5 2 lanes) to a target depth of 4.1 million read pairs. Two positive controls (ZymoBIOMICS Microbial Community Standard II, Log Distribution) and two negative controls (one pair per plate of extracted DNA) were sequenced alongside the biological samples, and provide no evidence of kit contamination (see supporting information). In addition to DNA from 83 samples extracted using the QIAamp 96 PowerFecal protocol described, we also used iGenomx Riptide protocols to prepare and sequence 13 of the same DNA extracts used to create the Nextera XT deeply sequenced dataset (see above). This allowed us to more directly compare the results generated using different library preparation techniques, by eliminating methodological and technical variation related to DNA extraction. Three samples from the Nextera XT prepared dataset were excluded from the iGenomx Riptide preparation and sequencing, because the original DNA extract was depleted.