Yes, many bacterial taxa are identified deeper than the family-level. In fact, the majority of the bacterial taxa are only a single strain because there’s not much known in the scientific literature about family membership.

Yes! You can read more about PhyloChip technology here.

The controls built into the assay address cross-hybridization. There are probe specific controls, controls for each probe called mismatch controls, as well as external controls that are spiked in to make the analysis robust.

The PhyloChip™ assay design and the analysis incorporate knowledge about chimeras to minimize the likelihood that they will cause a problem.

Yes!

Yes, Midgley et al. 2012 describes an in silico analysis of 16S rDNA probe sequences from the Greengenes website designed to evaluate the accuracy of microarray-based phylogenetic analysis. The authors claim that the probe sequences analyzed correspond to the G2 PhyloChip array probe sets and that prior analysis of G2 array data may have yielded an inflated number of detected taxa and distorted community structure.

It is clear, however, that the author's probe-to-OTU map is not correct and is not useful for interpreting G2 PhyloChip data. For the in silico microarray described in this paper, many more probes were assigned to probe-sets than were actually used in any G2 PhyloChip analysis. To our knowledge, there is no such direct probe set list available through PhyloTrac or Greengenes, as the authors claim.

The result of this incorrect probe assignment is that it leads to the erroneous conclusions published in this study. For example, in Figure 2, 135 probe sets were identified as part of their analysis. A majority of these probe sets, 118 out of 135 or 87% have incorrectly assigned probes. Based on our analysis, a total of 2,362 probes were incorrectly assigned to the 118 probe sets.

While the analysis in this paper is flawed, it is known that older probe set annotations of the G2 array OTUs contain some redundancy (on the order of 20 OTUs). However, Second Genome maintains and distributes updated annotation maps of the G2 and G3 OTUs in which this redundancy has been corrected. The authors of this publication were provided this updated G2 annotation map but did not appear to incorporate it into their analysis.

No less than 10 – 30 ng/μl.

Human DNA doesn’t cause cross hybridization problems because the probes have been selected to be specific to bacterial 16S rRNA genes. There could be an issue in the PCR step is if the relative amount of human to bacterial DNA causes inefficient amplification of bacterial DNA (i.e. the bacterial template is so low that the primers have trouble “finding” it.) There are many papers where samples containing human DNA were successfully analyzed for 16S content.

Lung tissue: Electrophoresis. 2010 Jul;31(14):2411-5.
Skin collections: J Med Dent Sci. 2010 Mar;57(1):65-74.
Liver tissue: Scand J Gastroenterol. 2010;45(2):160-7.
Skin wounds: Proc Natl Acad Sci U S A. 2010 Aug 17;107(33):14799-804. Epub 2010 Jul 28.

Our input DNA requirement is 50 ng total mass of good quality DNA from each sample for PCR. We request that you send it at a concentration of 10 ng/ul. We spilt each sample into 2x 25 ng aliquots, carry out the PCR in replicate and then pool, purify, concentrate, and quantify. We always ask that people send at least 200 ng of gDNA because of the vagaries of DNA quantification methods. The Picogreen method is preferred here because we find it is one of the more accurate methods for measuring double stranded PCR-able DNA.

4.2 fg/cell
Kubitschek, HE and Friedman, MC (1971) Chromosome replication and the division cycle of Escherichia coli B/r. Journal of Bacteriology. 107:95-99.

We often receive questions from researchers starting fecal microbiome studies about recommended procedures and expected yields for microbial DNA isolation, as well as proper storage, processing, and shipping conditions.
In our whitepaper, The Inside Scoop on Poop, we've tried to do the dirty work of gathering tips and tricks from leading experts in fecal DNA isolation and microbiome analysis so that you can get to the…um…dirtier work of starting your own microbiome study.

Yes, the PhyloChip assay can be used to analyze microbial communities from mouse samples as well as any other model organism or environment. The 16S rRNA sequences represented on the PhyloChip array are derived from the 16S rRNA gene database, greengenes.lbl.gov, which contains aligned, chimera screened, and taxonomically classified sequences imported from the NCBI GenBank sequence database. Sequences deposited in GenBank come from a huge variety of microbial sources, including humans, model organisms and environmental samples. These 16S rRNA sequences correspond to named or unnamed microbial isolates, an unnamed symbiont, and, very often, uncultured organisms.

Very low variation has been observed among replicate abundance scores for an OTU across arrays. In the earliest G3 paper (Hazen, 2010, Science), data was collected using a Latin Square experiment where artificial communities were mixed at 26 defined concentrations. Then, the 26 mixes were assayed on three separate days. The triplicate abundance scores from the same OTU measured on three different days produced an average coefficient of variation of 0.109. This is why microarray quantification of mixed nucleic assays has been employed by research teams requiring reproducible results.

Our project pricing depends on a few variables, including the type of DNA isolation you may need as well as the number of samples you plan to analyze. If you would like to speak with someone about your specific project, please complete this form and we'll be in contact with you shortly.

We have a minimum requirement of 10 samples. Collecting 5 samples from 2 conditions is perfectly acceptable. The thing to keep in mind is that you will be comparing bacterial populations between samples and therefore, the higher the number of samples from each condition you process, the stronger your comparison data will be between conditions. We recommend a minimum of 3 samples per condition, but 5 is better. That way if you have an outlier, it will not obliterate the results from the other samples. Just make sure that you're not taking all 5 samples from the same person/animal, etc. Choose different animals or people that are as closely matched as possible to serve as your baseline for comparison.