Still Not Use QIIME2 for Amplicon Analysis? You’re OUT!
What is QIIME 2?
The Quantitative Insights Into Microbial Ecology (QIIME) microbiome bioinformatics platform has supported many microbiome studies and gained a broad user and developer community.
QIIME 2™(https://qiime2.org) is a new version of the microbiome analysis platform compare to Qiime1.
Originally conceptualized and developed by Professor Gregory Caporaso of Northern Arizona University, QIIME was designed to take users from raw DNA sequence data and finish with publication-quality figures and statistical results (Caporaso et al. 2010). QIIME 1 is now deprecated and no longer in use – it has been rewritten into QIIME 2 in 2016 with 112 colleagues from 79 units around the world. The QIIME 2 paper was written in Python 3 and published online on July 24th (2019) in the world’s top academic Journal “Nature Biotechnology” (Bolyen et al. 2019).
As the analytical method that truly represents a golden standard for current (and future) analysis of microbiome Amplicon Sequencing data, it is no wonder that QIIME2 citations have been increasing since QIIME 2 became available online in 2018. According to Google Scholar statistics, the Bolyen et al. (2019) paper for QIIME2 has been cited over 2500 times in a period of only 2 years. As for QIIME 1, the Caporaso et al. (2010) paper has so far been cited over 25 000 times, which puts this open-source platform at the forefront of microbiome studies. Importantly though, the citations of the QIIME1 paper have more recently shown a downward trend as QIIME 1 is deprecated and no longer in use / being updated.
A novel functionality in QIIME2 (compared to QIIME 1) is a plug-in, which can be considered a standalone software project within the QIIME 2 pipeline. To perform analyses within QIIME 2, one can install different plugins that provide the specific analyses, depending on the pipeline we are aiming to develop. There are various QIIME 2 plugins that allow different analyses to be conducted, including: the q2-demux plug-ins (for demultiplexing samples & viewing sequence quality), q2-diversity plug-ins (to explore Alpha- and / or Beta-diversity metrices), and many others. Development / introduction of plug-ins into QIIME 2 allows the end-to-end microbiome analysis process, and importantly – users are provided with instructions to develop their own plug-ins and carry out additional analysis (when necessary)! This allows users to select and customize the analysis process for a specific need.
QIIME2’s classify-sklearn algorithm use Naive Bayes classifier, which has been documented to have a superior performance compared to other standard classification methods, particularly in terms of species annotation accuracy when using 16S rRNA gene and fungal ITS sequence datasets. In terms of the operating performance of the classifier, the computation time to carry out annotation analyses is more optimal than that of Vsearch, BLAST and other sequence annotation methods – as the number of our target / query sequences increases. QIIME 2 also outperforms other methods (such as RDP and SortMeRnA) as the reference sequence database increases, in terms of the operation time needed to conduct the taxonomic annotation. Overall, the Naive Bayes classifier (utilised in QIIME 2) outperforms other common classifiers in terms of operational stability and speed (Bokulich et al. 2018).
In addition to displaying statistical tables and static pictures in QIIME2, the process automatically provides a variety of interactive visual display forms, such as histograms of relative abundance of species, α-diversity, PCoA, etc. You can adjust the image format directly in the report, and output satisfactory images instantly.
Another specificity are the QZA and QZV files – these file formats are unique to QIIME 2. Apart from the actual data found within the QZA files, these files also contain information on the previous analysis steps undertaken, such as: the commands used, and the data that has been analysed. This allows to maximize the traceability and repeatability of the analysis steps, and additionally, the visualization results are included within the QZV files, such as: statistical tables, high-resolution graphs / plots, interactive web pages, and combined visuals. You can also upload QZA/QZV files online without installing the QIIME2 program, display the result chart in real time, and also simultaneously view the data analysis process, which greatly facilitates the sharing of analysis results.
Bolyen, E., Rideout, J.R., Dillon, M.R. et al. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat Biotechnol 37, 852–857 (2019). https://doi.org/10.1038/s41587-019-0209-9
Caporaso, J., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F., Costello, E., Fierer, N., Peña, A., Goodrich, J., Gordon, J., Huttley, G., Kelley, S., Knights, D., Koenig, J., Ley, R., Lozupone, C., McDonald, D., Muegge, B., Pirrung, M., Reeder, J., Sevinsky, J., Turnbaugh, P., Walters, W., Widmann, J., Yatsunenko, T., Zaneveld, J. and Knight, R., 2010. QIIME allows analysis of high-throughput community sequencing data. Nature Methods, 7(5), pp.335-336.
Bokulich, N., Kaehler, B., Rideout, J., Dillon, M., Bolyen, E., Knight, R., Huttley, G. and Gregory Caporaso, J., 2018. Optimizing taxonomic classification of marker-gene amplicon sequences with QIIME 2’s q2-feature-classifier plugin. Microbiome, 6(1).