Get Taxonomy Hierarchy using ETE Toolkit 3 and Apply to GenBank Assembly Summary

get_taxonomy_hierarchy_ete3.py

# This is roughly 30 fold faster than my original implimentation.
# Install ETE3 via pip and run ncbi.update_taxonomy_database() first.

from collections import defaultdict
import pandas as pd
from ete3 import NCBITaxa

ncbi = NCBITaxa()

def get_taxonomy_hierarchy(taxid):
    # Note that the resulting dictionary isn't ordered by the hierarchy.
    # Since there are missing levels for some taxids, the dict is converted to defaultdict,
    # which can be an unnecessary step in many cases.
    
    names = ncbi.get_taxid_translator(ncbi.get_lineage(taxid))
    ranks = ncbi.get_rank(ncbi.get_lineage(taxid))
    return defaultdict(lambda: "", {ranks[k]: names[k] for k in names.keys()})


# Let's assign taxonomic classifications for GenBank entries.
df = pd.read_csv("assembly_summary.txt", skiprows=1, sep="\t")
df["tax_hierarchy"] = df["species_taxid"].apply(lambda x: get_taxonomy_hierarchy(x))

for rank in ["species", "genus", "family", "order", "class", "phylum"]:
    df[rank] = df["tax_hierarchy"].apply(lambda x: x[rank])