PPIT: an R package for inferring microbial taxonomy from nifH sequences

Area of Science:

• Environmental microbiology
• Metagenomics
• Bioinformatics

Background:

• Linking microbial taxa to ecosystem functions is crucial in environmental microbiology.
• Inferring taxonomy from marker genes like nitrogenase reductase (nifH) is challenging due to horizontal gene transfer.
• Existing methods for nifH taxonomy are difficult to scale and prone to errors.

Purpose of the Study:

• To develop a novel R package, Phylogenetic Placement for Inferring Taxonomy (PPIT), for accurate microbial taxonomy assignment from nifH amplicon sequences.
• To address the limitations of existing methods in handling horizontal gene transfer events in nifH evolution.
• To provide a scalable and standardized approach for taxonomic profiling of microbial communities.

Main Methods:

• PPIT utilizes phylogenetic placement on a reference nifH gene tree.
• It infers taxonomy by analyzing the phylogenetic neighborhood of query sequences.
• Inferences require taxonomic consistency and sufficient sequence identity to avoid errors from horizontal gene transfer.

Main Results:

• PPIT demonstrates higher accuracy in taxonomic inferences compared to BLAST-based methods.
• The package successfully identified Deltaproteobacteria as abundant potential diazotrophs in deep-sea sediment samples.
• Manual refinement of PPIT inferences achieved near-complete taxonomic assignment for the analyzed dataset.

Conclusions:

• PPIT offers a robust method for accurate microbial taxonomy assignment from nifH sequences.
• The package can be adapted for analyzing other marker genes in microbial ecology.
• Accurate taxonomic profiling is essential for understanding microbial community functions and ecological roles.