ITSoneDB_qiime

ITSoneDB is a curated database [1] collecting the Internal transcribed spacer 1 regions of the Eukaryotic ribosomal gene clusters.
This repository contains the resources for the implemetation of ITSoneDB in the QIIME2 pipeline [2].

Sequence clustering

ITS1 sequences available in the ITSoneDB were collected in the ITSoneDB_total_fasta_rel138.fa fasta file and clustered at 99% of similarity by using VSEARCH [3].

vsearch --cluster_fast ITSoneDB_total_fasta_rel138.fa -uc vsearch_cluster_99 \
--centroids 99_ref_ITSoneDB.fa --id 0.99

Taxonomic annotation

Figure1: A representation of a list of nodes belonging to a cluster in a reference taxonomy.

The obtained cluters representative sequences were taxonomically labelled by using two approaches:

Lowest common ancestor (LCA): this algorithm assigne a taxonomica label by considerign the lowest taxonomic rank common to all the sequences belonging to a cluster. Considering the Figure 1 we may assume that a hypothetical clusters contains the sequences belonging to the nodes M1, M2, M3 and M4 but not the sequences belongign to node N1. In this case the LCA node is the Ti node. The taxonomy file was generated by using the Python script common_taxonomy.py;
TANGO algorithm: this algorithm finds the most likely node by inferrig F-measures, the harmonic mean of Precision and Recall [4]. Let's considering the Figure 1 TANGO assignes the taxonomy to the node j2 because it is the one obtaining the best F-measure value. The taxonomy file was generated by using the Python script assign_taxonomy_with_tango.py.

Data

Data available in this repository:

ITSoneDB_total_fasta_rel138.fa.gz: compressed fasta file containing the ITSoneDB ITS1 sequences (use the link to download the file);
99_ref_ITSoneDB.fa.gz: ITS1 clusters representative sequences;
tax_99_ref_ITSoneDB.txt.gz: NCBI taxonomic path associated to the clusters representative sequences inferred by using the LCA algorithm;
tax_99_ref_ITSoneDB_tango_based.txt.gz: NCBI taxonomic path associated to the clusters representative sequences inferred by using the TANGO algorithm;

QIIME2 import commands

Following are listed the QIIME2 command to import ITSoneDB data as qzaartifacts:

qiime tools import \
  --input-path 99_ref_ITSoneDB.fasta \
  --output-path 99_ref_ITSoneDB.qza \
  --type 'FeatureData[Sequence]'

qiime tools import \
 --type FeatureData[Taxonomy] \
 --input-path tax_99_ref_ITSoneDB.txt \
 --input-format HeaderlessTSVTaxonomyFormat \
 --output-path ITSoneDB_taxonomy.qza

bfosso/ITSoneDB_qiime

ITSoneDB_qiime

Sequence clustering

Taxonomic annotation

Data

QIIME2 import commands

References