Table 1.
Percentages of predicted sequences assigned to second tier KO categories in the metagenomic dataset.

Thick lines separate tier 1 KO categories.
*Functional categories for which no reads were assigned are omitted.
†Predicted function only assigned at the first tier.