Computational protocol: The Effect of Root Exudate 7,4′-Dihydroxyflavone and Naringenin on Soil Bacterial Community Structure

Similar protocols

Protocol publication

[…] DNA was extracted from 250 mg of soil from each sample with Power Soil DNA Isolation Kit (MO BIO Laboratories, USA). The V4 region of the 16S gene was amplified with PCR using the primers of Kozich et al. []. The reactions contained 22.5 μl AccuPrime Pfx SuperMix (Invitrogen, USA), 7.5 ng DNA, and 7.5 pmol forward and reverse primers (Integrated DNA Technologies, USA) in 25 μl final volume. Amplification was carried out in a Bio-Rad My Cycler version 1.065 thermocycler (Bio-Rad, USA) with 4 min initial denaturation at 95°C followed by 30 cycles of 20 s at 95°C, 15 s at 55°C, and 2 min at 68°C, and a final extension for 10 min at 68°C. The size and quality of the PCR products were checked using agarose gel electrophoresis. The PCR products were cleaned, quantified, pooled, and sequenced on an Illumina MiSeq instrument with a 500 cycle v2 kit (Illumina, USA) at the University of Kentucky Advanced Genetics Technologies Center according to the protocol of Kozich et al. []. Samples from the 7,4′-dihydroxyflavone and the naringenin experiments were sequenced separately. The sequence data is accessible at the National Center for Biotechnology Information Sequence Read Archive ( under the accession number PRJNA295777.Forward and reverse sequence reads were joined, and then sequences that were low quality, chimeric, mitochondrial, chloroplast, archaeal, eukaryotic, and unclassifiable were removed in mothur v1.34 [] as described in the MiSeq SOP (, accessed December 2014) using the SILVA alignment [] release 119, and the Ribosomal Database Project [] release 10. Sequences were binned to operational taxonomic units (OTUs) using minimum entropy decomposition (MED) [] with the following settings: m = 0.0965, c = 4, M = (number of sequences in the dataset/10 000), and V = 3. A representative sequence from each OTU was classified according to the Ribosomal Database Project release 10 using mothur with 70% bootstrap cutoff.Instead of rarefying [], centered log-ratio (CLR) transformation [,] was applied to the data matrices using the compositions package [] in R version 3.2.1 ( The data matrices from the 7,4′-dihydroxyflavone and naringenin experiments contained 13 and 43 zeroes, respectively. These were replaced with ones to allow this transformation. Ordination plots were made with non-metric multidimensional scaling (NMS) in PC-ORD 6.0 (MJM Software Design, USA) with Euclidean distances: 250 runs were performed with random starting configurations in one to six dimensions with a 10−7 instability criterion and 500 maximum iterations with 0.2 initial step length to find the best starting configurations in each dimensionality. Statistics for the final stress for each dimensionality were obtained from 250 runs with randomized data. Dimensions were only accepted if they decreased the stress to a lower value than that from 95% of the randomized runs. Based on the results, two-dimensional solutions were selected. The final run was conducted using the determined best starting configuration. Multi-response permutation procedure (MRPP) [] was used on the CLR transformed datasets with Euclidean distances to test the significance of the differences between the high, medium, low, and control treatments. To find differentially abundant OTUs in the treatments, we used DESeq2 (without CLR transformation) [] as recommended by McMurdie et al. [] with the DESeq2 package version 1.8.1 [] in R. To account for the high number of simultaneous tests, q-values were calculated according to the positive false discovery rate method [] with the smoother option in QVALUE ( The q-value of a test estimates the proportion of false positive findings in the dataset if tests with equal or lower p-values are accepted as significant. […]

Pipeline specifications

Software tools mothur, DESeq2
Databases SRA
Application 16S rRNA-seq analysis
Organisms Bacteria, Arabidopsis thaliana
Chemicals Flavonoids