Home

Using LIN-taxonomy for Metagenomics classification

This tool creates LIN (Life identification number) based taxonomy files for any given list of input genomes. Instead of traditional ranks, LIN-taxonomy is based on 20 thresholds/ranks of average nucleotide identity (ANI). These thresholds of similarity ranges from 70%-99.999%. The current default implementation of LIN thresholds are outlined below:

Traditional taxonomy considers 'species' as the lowest rank which corresponds to approximately 95% ANI. However, with LIN-taxonomy you get much more resolution and 14 more ranks at the sub-species level which allows for accurate outbreak level detection of pathogens.

Adaptation with Metagenomics classifiers

LIN-taxonomy tool is implemented to create NCBI-like taxonomy files that can be used with any Metagenomics classification tool that allows to build custom taxonomy and databases. So far, this has been tested with the following:

• kraken2
• krakenUniq

Click here for LIN-taxonomy implementation with Sourmash.