This is my attempt to collate the literature on how easy it is to introduce bias into microbiome studies.  I hope to crowd-source papers and add them below under each category.  PLEASE GET INVOLVED.  If this works well we can turn it into a comprehensive review and publish 🙂 Add new papers in the comments or Tweet me 🙂

Special mention to these blogs:

  1. Microbiomdigest’s page on Sample Storage
  2. Microbe.net’s Best practices for sample processing and storage prior to microbiome DNA analysis freeze? buffer? process?
  3. The-Scientist.com’s Spoiler Alert

****************************************

UPDATE 7th August 2016

****************************************

For prosperity the original blog post is below, but I am now trying to manage this through a ZOTERO GROUP LIBRARY.  Please continue to contribute – join the group, get involved!  I think you can join the group here.  I am trying to avoid just listing software papers BTW I would prefer to focus on papers that specifically demonstrate sources of bias.

I won’t be updating the text below.


General

  • “We propose that best practice should include the use of a proofreading polymerase and a highly processive polymerase, that sequencing primers that overlap with amplification primers should not be used, that template concentration should be optimized, and that the number of PCR cycles should be minimized.” (doi:10.1038/nbt.3601)

Sample collection and sample storage

  • “Samples frozen with and without glycerol as cryoprotectant indicated a major loss of Bacteroidetes in unprotected samples” (24161897)
  • “No significant differences occurred in the culture-based analysis between the fresh, snap or -80°C frozen samples.” (25748176)
  • “In seven of nine cases, the Firmicutes toBacteroidetes 16S rRNA gene ratio was significantly higher in fecal samples that had been frozen compared to identical samples that had not….. The results demonstrate that storage conditions of fecal samples may adversely affect the determined Firmicutes to Bacteroidetes ratio, which is a frequently used biomarker in gut microbiology.” (22325006)
  • “Our results indicate that environmental factors and biases in molecular techniques likely confer greater amounts of variation to microbial communities than do differences in short-term storage conditions, including storage for up to 2 weeks at room temperature” (10.1111/j.1574-6968.2010.01965.x)
  • “The trichloroacetic acid preserved sample showed significant loss of protein band integrity on the SDS-PAGE gel. The RNAlaterpreserved sample showed the highest number of protein identifications (103% relative to the control; 520 ± 31 identifications in RNAlater versus 504 ± 4 in the control), equivalent to the frozen control. Relative abundances of individual proteins in the RNAlater treatment were quite similar to that of the frozen control (average ratio of 1.01 ± 0.27 for the 50 most abundant proteins), while the SDS-extraction buffer, ethanol, and B-PER all showed significant decreases in both number of identifications and relative abundances of individual proteins.” (10.3389/fmicb.2011.00215)
  • “Optimized Cryopreservation of Mixed Microbial Communities for Conserved Functionality and Diversity” (10.1371/journal.pone.0099517)
  • “A previously known bias in FTA(®) cards that results in lower recovery of pure cultures of Gram-positive bacteria was also detected in mixed community samples. There appears to be a uniform bias across all five preservation methods against microorganisms with high G + C DNA. Overall, the liquid-based preservatives (DNAgard(™), RNAlater(®), and DESS) outperformed the card-based methods.” (22974342)
  • “Microbial composition of frozen and ethanol samples were most similar to fresh samples. FTA card and RNAlater-preserved samples had the least similar microbial composition and abundance compared to fresh samples.” (10.1016/j.mimet.2015.03.021)
  • “Bray-Curtis dissimilarity and (un)weighted UniFrac showed a significant higher distance between fecal swabs and -80°C versus the other methods and -80°C samples (p<0.009). The relative abundance of Ruminococcus and Enterobacteriaceae did not differ between the storage methods versus -80°C, but was higher in fecal swabs (p<0.05)” (10.1371/journal.pone.0126685)
  • “We experimentally determined that the bacterial taxa varied with room temperature storage beyond 15 minutes and beyond three days storage in a domestic frost-free freezer. While freeze thawing only had an effect on bacterial taxa abundance beyond four cycles, the use of samples stored in RNAlater should be avoided as overall DNA yields were reduced as well as the detection of bacterial taxa.” (10.1371/journal.pone.0134802)
  • “A key assumption in many studies is the stability of samples stored long term at −80 °C prior to extraction. After 2 years, we see relatively few changes: increased abundances of lactobacilli and bacilli and a reduction in the overall OTU count. Where samples cannot be frozen, we find that storing samples at room temperature does lead to significant changes in the microbial community after 2 days.” (10.1186/s40168-016-0186-x)

DNA extraction

  • “Caution should be paid when the intention is to pool and analyse samples or data from studies which have used different DNA extraction methods.” (27456340)
  • “Samples clustered according to the type of extracted DNA due to considerable differences between iDNA and eDNA bacterial profiles, while storage temperature and cryoprotectants additives had little effect on sample clustering” (24125910)
  • “Bifidobacteria were only well represented among amplified 16S rRNA gene sequences when mechanical disruption (bead-beating) procedures for DNA extraction were employed together with optimised “universal” PCR primers” (26120470)
  • Qiagen DNA stool kit is biased (misses biffidobacteria)  (26120470)
  • “Bead-beating has a major impact on the determined composition of the human stool microbiota.  Different bead-beating instruments from the same producer gave a 3-fold difference in the Bacteroidetes to Firmicutes ratio” (10.1016/j.mimet.2016.08.005)
  • “We observed that using different DNA extraction kits can produce dramatically different results but bias is introduced regardless of the choice of kit.” (10.1186/s12866-015-0351-6)

Sequencing strategy

  • Bifidobacteria were only well represented among amplified 16S rRNA gene sequences … with optimised “universal” PCR primers. These primers incorporate degenerate bases at positions where mismatches to bifidobacteria and other bacterial taxa occur” (26120470)
  • Anything other than 2x250bp sequencing of V4 region (approx 250bp in length) inflates number of OTUs (23793624)
  • “This study demonstrates the potential for differential bias in bacterial community profiles resulting from the choice of sequencing platform alone.” (10.1128/AEM.02206-14)
  • “The effects of DNA extraction and PCR amplification for our protocols were much larger than those due to sequencing and classification” (10.1186/s12866-015-0351-6)
  • “Nested PCR introduced bias in estimated diversity and community structure. The bias was more significant for communities with relatively higher diversity and when more cycles were applied in the first round of PCR” (10.1371/journal.pone.0132253)
  • “pyrosequencing errors can lead to artificial inflation of diversity estimates” (19725865)
  • “Our findings suggest that when alternative sequencing approaches are used for microbial molecular profiling they can perform with good reproducibility, but care should be taken when comparing small differences between distinct methods” (25421243)

Data analysis strategy

 

Bioinformatics / database issues

 

Contamination in kits

  • “Reagent and laboratory contamination can critically impact sequence-based microbiome analyses” (25387460)
  • “Due to contamination of DNA extraction reagents, false-positive results can occur when applying broad-range real-time PCR based on bacterial 16S rDNA” (15722157)
  • “Relatively high initial densities of planktonic bacteria (10(2) to 10(3) bacteria per ml) were seen within [operating ultrapure water treatment systems intended for laboratory use]” (8517737)
  • “Sensitive, real-time PCR detects low-levels of contamination by Legionella pneumophila in commercial reagents” (16632318)
  • “Taq polymerase contains bacterial DNA of unknown origin” (2087233)

 

Commercial stuff

 

 

 

 

 

Bibliography