# Bioconductor Biomformat

> This is an R package for interfacing with the BIOM format. This package includes basic tools for reading biom-format files, accessing and subsetting data tables from a biom object (which is more complex than a single table), as well as limi

- Skill: `biomate-ai/bioconductor-biomformat` (Agent Skill)
- Install (CLI): `npx skillmds@latest add biomate-ai/bioconductor-biomformat`
- Raw SKILL.md: https://api.skillmd.com/api/skills/biomate-ai/bioconductor-biomformat/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: biomate-ai (https://skillmd.com/u/biomate-ai)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/biomate-ai/bioconductor-biomformat

---


# biomformat

## Dependencies & Environment

> Package-intrinsic requirements from the Bioconductor landing page — reproduce in any R environment.

- **Version:** 1.40.0 · **Bioconductor:** 3.23 · **R:** ≥ 4.6
- **Imports:** jsonlite, Matrix
- **Install:** `BiocManager::install("biomformat")`

## When to Use
- Reading BIOM format files (JSON or HDF5) into R using `read_biom()`.
- Extracting core observation data (OTU tables) as sparse matrices using `biom_data()`.
- Converting BIOM objects to `SummarizedExperiment` objects using `biom_to_SummarizedExperiment()`.
- Constructing a BIOM object from standard R matrices and data frames using `make_biom()`.
- Converting BIOM data into a tidy long-format data frame for downstream analysis using `as_tibble.biom()`.

## When NOT to Use
- For statistical analysis of microbiome data, use `phyloseq` or `DESeq2` because `biomformat` is strictly an I/O and utility package.
- For raw sequence processing, use `dada2` because `biomformat` operates on already-constructed contingency tables.

## Data Requirements
- BIOM format files (v1 JSON or v2 HDF5).
- For manual construction via `make_biom()`, a count matrix (features in rows, samples in columns) and optional `data.frame` objects for observation and sample metadata.

## Key Parameters
- **data** (default): The count matrix when using `make_biom()`.
- **observation_metadata** (default): Taxonomy or feature metadata `data.frame` when using `make_biom()`.
- **matrix_element_type** ("int"): The data type of the matrix elements in `make_biom()`.
- **rows** (default): Character vector to subset features directly in `biom_data()`.
- **columns** (default): Character vector to subset samples directly in `biom_data()`.

## Best Practices
- Use `read_biom()` which automatically detects and routes HDF5 files to `read_hdf5_biom()`.
- Use `as_tibble.biom()` to convert data into a tidy long-format for downstream `dplyr` and `purrr` workflows.
- For large datasets (>2GB), use `write_hdf5_biom()` instead of `write_biom()` to avoid JSON serialization limits.

## Common Pitfalls
- Attempting to print a massive BIOM object streams too much data; fix this by relying on the brief summary printed by default or using `biom_data()` to inspect subsets.
- Functions expecting standard matrices fail on sparse matrix classes; fix this by coercing with `as(biom_data(x), "matrix")`.
- Writing very large tables with `write_biom()` fails due to JSON string limits; fix this by using `write_hdf5_biom()` for HDF5 format.

## Alternatives
- `phyloseq`: Has many more utilities for interacting with and analyzing this kind of data.
- `SummarizedExperiment`: The standard Bioconductor container for rectangular feature-by-sample assay data (though `biomformat` interoperates with it).

## Citations
- Paul J. McMurdie and Joseph N Paulson (2015). biomformat: An interface package for the BIOM file format. R/Bioconductor package version 1.0.0.

## References
- Homepage: https://bioconductor.org/packages/biomformat
- Vignette: https://bioconductor.org/packages/release/bioc/vignettes/biomformat/inst/doc/biomformat.html

<!-- biomate-cta -->
---

## Run this on BioMate

This skill is the **knowledge layer** — when, why, and how to use `biomformat`. To **run this analysis on your own data** with managed compute, automated QC, and reproducible outputs, use **[BioMate](https://www.biomate.ai?ref=kb&pkg=biomformat)** — free to start.

▶ **[Open `biomformat` on BioMate →](https://www.biomate.ai?ref=kb&pkg=biomformat)**