AxioParse: streamlining Axiom Microbiome assay data processing and dataset generation

The Applied Biosystems Axiom Microbiome Array enables high-throughput detection of bacteria, archaea, viruses, protozoa, and fungi across multiple samples. However, its native software outputs are not compatible with common downstream analysis tools, requiring preprocessing. We identified a lack of open-source pipelines tailored to these outputs. To address this gap, we developed AxioParse, a Python-based pipeline built with the Dagster orchestration framework that automates data cleaning, taxonomic mapping, and formatting for downstream analysis. AxioParse reduces manual processing and generates datasets compatible with platforms such as QIIME2 and R, improving reproducibility and facilitating broader use of the Axiom Microbiome Array in microbiome research (https://github.com/Eghtesady-Lab-Bioinformatics/axioparse).
The Axiom Microbiome Array detects diverse microbial taxa across multiple samples, but its native outputs require substantial data wrangling before analysis in platforms such as R or QIIME2. We developed AxioParse, a Python pipeline that automates data cleaning, taxonomy updates, and formatting for downstream tools. The pipeline is customizable and supported by documentation and example datasets on GitHub.
AxioParse is an open-source Python pipeline built with Dagster that automates ingestion and processing of Axiom Microbiome Array outputs. The pipeline consolidates probe-level MiDAS data into species-level tables, updates taxonomy using NCBI Entrez, and exports formatted datasets for downstream analysis.