More posts like this
-
hclust2.py heatmap in MetaPhlAn3 --ftop features
written by serene.s •
-
UMAP of 10X and SmartSeq2 datasets on same plot
written by a.rex
-
Trying to understand samtools -f and -F
written by shibl_a •
-
Best practices to perform a gene set enrichment analysis (GSEA) with E. coli?
written by Matteo Schiavinato
-
Sample level integration of two RNASeq datasets
written by Kristoffer Vitting-Seerup
-
Identifying shared exact sequence variants between two datasets
written by adam.sorbie •
-
crossing multiple annotations and validation with RNA-seq
written by guillaume.rbt
-
Metagenomic Iontorrent reads: filtering and denoising
written by fhsantanna
-
How compare two de novo assembled transcriptomes of the same sample with different sequencing depth?
written by seta
-
How To Retrieve An Ancestral Sequences In A Phylogenetic Tree?
written by spandankeshav •
I think you can not rely on a single tool, in this case, I would suggest you can use multiple tools and check the results and then decide. Because the problem for all these tools (
Kaiju, MetaPhlAn2 or Kraken) is, all of them will assign taxonomy on the basis of reference dataset.I think MetaPhlAn2 will use their own marker gene set whereas Kraken and Kaiju will use
NCBI nrdb(kaiju = protein nrdbandKraken nucleotide nrdb).In this case, as your samples are from seawater, then either Kraken or kaiju would be a good choice, because of the reference database(I think).