Sequencing on the AZD0156 cell line genome Sequencer FLX platform The PCR products were processed for parallel-tagged sequencing on the Genome Sequencer FLX platform, as described elsewhere [38]. Briefly, sample-specific barcode sequences were ligated to the PCR products, and DNA concentrations were assessed with a Mx3005P™ qPCR System (Stratagene). Samples were then pooled in equimolar ratios to a total Selleckchem Apoptosis Compound Library DNA amount of 440 ng. The pooled
DNA was subsequently amplified in PCR-mixture-in-oil emulsions and sequenced on a Genome Sequencer FLX /454 Life Sciences sequencer (Branford CT), according to the manufacturer’s protocol. Data analysis The initial sequence reads were filtered to remove low-quality sequences and artifactual sequence reads (i.e., reads containing two or more different tags, no tags, primers in selleck chemicals llc the middle of sequence reads, or lacking a primer sequence). After removing sequences less than 200 bp in length (as these may not give reliable results), there were 48,168 sequence reads used in the analysis. These sequence reads have been deposited in GenbankSequence Read Archive (SRA) SRP015938. A genus was assigned to each sequence by comparing the filtered sequences against the Ribosomal
Database Project [16] using the online program SEQMATCH (http://rdp.cme.msu.edu/seqmatch/seqmatch_intro.jsp) and a threshold setting of 90%. Diversity statistics and the apportionment of variation based on the frequency distribution of genera within and between individuals were calculated with the Arlequin 3.1 software [39]. Spearman’s rank correlation coefficients, sharing (Venn) diagrams, and Analysis of Similarity (ANOSIM) [40] were calculated with the R package. Rarefaction analysis was carried out using the Resampling Rarefaction 1.3 software ADAMTS5 (http://strata.uga.edu/software/). Partial correlation
analysis was carried out with the GeneNet package [41]. For the UniFrac analysis, the sequences were aligned with the Infernal 1.0 program [42] and a phylogenetic tree was constructed under a generalized time reversible (GTR) model with the FastTree software [43]. Fast UniFrac [19] was then used to compare the microbial communities, compute the distance matrix, and generate the cluster tree. The phylogenetic tree from FastTree was also used to calculate Faith’s Phylogenetic Diversity [20] using the “picante” package in R [44]. The OTU networks were constructed from the sequences aligned with Infernal 1.0 by using tools provided by the RDP website to first cluster all sequences that were 97% or more similar (based on a minimum overlap of 25 bases) into OTUs (to account for sequencing errors). We then used the Cytoscape 2.8 software [45] to generate and visualize the networks. Briefly, each individual is considered a Source node and each OTU is a Target node.