Metagenomic tools comparison on 16S rRNA Sequences
| Main Authors: | Talavera Rodríguez, Alba María, Marcos-Zambrano, Laura Judith, Carrilo de Santa Pau, Enrique, Serrano Villar, Sergio, Fernandez Lanza, Val |
|---|---|
| Format: | Proceeding poster Journal |
| Bahasa: | eng |
| Terbitan: |
, 2019
|
| Subjects: | |
| Online Access: |
https://zenodo.org/record/4518945 |
Daftar Isi:
- Background: Metagenomics, the study of genomic sequences present in a given environment or ecosystem, has become an increasingly popular field of study during the last years. As a consequence, there are a huge amounts of metagenomics data and tools available but there is no studies that compare the results of using different bioinformatics tools for 16S rRNA sequences. Kraken2 is an alternative to well established methods such us QIIME2 or Mothur. Kraken2 is based on exact k-mer matches to achieve high accuracy and fast classification meanwhile QIIME and Mothur are based on clustering and classification. New version of Kraken2 allow to use 16S rRNA databases as reference. Here we evaluate the three different tools in order to determine how similar are theirs results and try to clarify something about what should be the standard method for develop the metagenomics analyses. Materials/methods: We used a well-kown dataset available at Mothur’s tutorial, which comes from a study interested in understanding the effect of normal variation in the gut microbiome on host health using mice fresh feces. This sample consist on 16S rRNA sequencing that were generated using Illumina's MiSeq platform using paired end reads. In order to compare the results from the three methods we use GreenGenes database to classification. We have used Qiime 2 (qiime2-2019.1), Kraken 2 (version 2.0.7-beta) and Mothur (v.1.40.5) because they are becoming popular in metagenomics analysis. The taxonomic classification results were evaluated with different methods at all the taxonomic levels to determine how different are their classifications and if their results are comparable. Results: The results show that Mothur are really different in comparison with the other tools, but results from Kraken 2 and Qiime 2 are practically identical at all the taxonomic levels, except at gender level that is a bit different and at specie level that is really different using the three tools. Conclusions: The results suggest that Kraken2 can be used for target metagenomics analysis as substitute of more classical methods such as QIIME2 or Mothur. Moreover Kraken2 is computationally much faster and simple to use.