What it is
Canonical name: Kraken2. 1
Known aliases: No aliases were reported in the source document. 1
Stable external identifier: None was reported in the source document. 1
Kraken2 is a taxonomic classification method whose NMDC deployment used a prokaryote-restricted reference database. 1
Evidence from euk_in_prok_correlates
Kraken2 produced approximately 0 domain-level Eukaryota signal in the analysis of 2,759 NMDC ReadbasedAnalysis runs because its NMDC reference database was prokaryote-restricted. 1
The only eukaryotic kingdom represented in the Kraken2 deployment was Metazoa/human. 1
Because Kraken2 produced approximately 0 usable domain-level Eukaryota signal, it was not used to estimate the eukaryotic fraction in this analysis. 1
The result demonstrates that Kraken2 and centrifuge were not interchangeable with gottcha2 for eukaryote quantification in this NMDC deployment, because GOTTCHA2 was plastid- and eukaryote-aware whereas the Kraken2 and Centrifuge databases were prokaryote-restricted. 1
Accordingly, cross-collection contamination-quality-control analyses should account for classifier and database compatibility rather than treating Kraken2-derived and GOTTCHA2-derived eukaryotic fractions as equivalent. 1
Evidence from harvard_forest_warming
The Harvard Forest study refines the eukaryote-detection limitation: despite the prokaryote-restricted deployment, Kraken2 read-based relative abundance was used for bacterial community comparison and detected Actinobacteria increasing from 0.249 to 0.315 and Acidobacteria decreasing from 0.035 to 0.024 in heated organic soil, with q=0.049 for both changes. 2
This supports using Kraken2 for prokaryotic community-level contrasts while retaining the existing warning that its output is unsuitable for estimating eukaryotic fractions in this deployment. 2,1
This method-specific limitation is relevant to multi-omics-integration, which concerns integration of taxonomic results with environmental and sequencing metadata, and to cross-tenant-data-bridging, which covers linking NMDC workflow results with sample metadata. 1
See the project summaries: Metadata Correlates of Eukaryotic Contamination in NMDC Prokaryote-Targeted Metagenomes and Harvard Forest Long-Term Warming — DNA vs RNA Functional Response. 1,2