Following my previous question on metagenomics (see link) I was wondering if replicate samples are necessary or convenient for recording changes on the community functioning over time. I search for many papers on metagenomics and I couldn't find any study where authors assess replicate samples.
I want to analyse both the microbial functioning and the diversity. I know that from metagenomic data I can obtain the 16S/18S suitable for addressing the community diversity (tagged Illumina) but in case I don't have/need to do metagenomic replicates (much more expensive), I guess I may need replicates for the tagged 16S/18S data...
By the way, can I really use metagenomics to evaluate community changes? I know that it basically gives you the metabolic capabilities of the community rather than the metabolic activity at a given time, but could I use the coverage or some other technique to assess changes without having to do metatranscriptomics or metaproteomics?
Thank you in advance.
https://www.researchgate.net/post/Should_the_eukaryotes_be_filtered_out_in_a_metagenomic_study_of_a_radioactive_waste_site?_tpcectx=qa_overview_asked&_trid=552de8c5d3df3e79548b4598_
As previous answers, you MUST HAVE replicates, and the article from Prosser (2010) clearly answer your question: "replicate or lie" regardless the techniques you use. It is not preferable, it is a duty to have replicates and unfortunately some articles still published without proper replication when using metagenomic. As a reviewer, I reject articles without replication if metagenomics are the main results, or ask to remove those data or mentioned the results briefly as no conclusions can be drawn from non-replicated analysis...
The article from Prosser should be read by every researcher...
To answer the second part of your question, yes metagenomic can give you information about change over time. You might find the article from Fierer and Ladau (2012) useful: "Predicting microbial distribution in space and time".
Thank you all. I think now my biggest doubt would be how to treat the replicates, how to analyse them properly: statistical analyses and such. Anybody has experience on that.
Again thank you.
There is no perfect answer to your question as its depends of your experiment, research question and statistical analysis. So the number of replicates should be calculated based on these parameters, as you would do for any experiment/analysis. The amount of money you have will also determine the number of replicates you can analyse… but at least 3 replicates should be used. However, when you work with environmental samples from a field experiment from example, 3 replicates are often not enough to obtain clear results and more samples should be used, but again there are no absolute rules. You will see studies with a large number of replicates or samples used to reveal trend in metagenomics data. So nobody can answer your question without knowing the research question you are trying to answer.
I totally agree with Aimeric, replicate or lie is one of my favorite yet inspiring articles. More replicates is better your analysis and your understanding of the the changes in your communities
Good luck
Ahmed
Another article from Prosser just released, which is also inspiring (in my opinion) and rise up some important points, limitations and questions about “omics” methods, and the chase of the latest methods rather than the biological questions…
Prosser, J.I., 2015. Dispersing misconceptions and identifying opportunities for the use of “omics” in soil microbial ecology. Nature Reviews Microbiology advance online publication. doi:10.1038/nrmicro3468
Thank you very much for your valuable contributions. I wonder if it would be possible to pool the replicas of a sample into one run?
If not, How can I analyze results for each sample with replicas ?
Abdelmalek Lekired I think, pool sample from one field and then make replicates from that pooled sample, In this way, result will really depict the scenario of one field, or from four fields, each field has four replicates, so four field four treatment
- Badges
- Science topic
- Similar topics
- Geoscience
- Atmospheric Sciences
- Climate Science
- Climate Change