Install alevin-fry with Homebrew

Project links

• Homepage

Commands and aliases

• alevin-fry

Project history

The official README says alevin-fry consumes RAD files produced by piscem or salmon alevin, generates permit lists, and estimates distinct molecules per gene per cell. The project focuses on safety, accuracy, time efficiency, and memory efficiency.

The README and documentation present alevin-fry as the successor to alevin. It subsumes core alevin features, adds capabilities, improves performance, and is where the maintainers expect most future method development to happen, while salmon alevin remains maintained for users not ready to migrate.

The project was described in the 2022 Nature Methods paper 'Alevin-fry unlocks rapid, accurate and memory-frugal quantification of single-cell RNA-seq data.' Its changelog shows active releases through the 0.9 series, including USA-mode support, UMI resolution modes, dependency updates, and command-line validation improvements.

Adoption history

The official README documents Bioconda availability for x86 Linux and macOS, crates.io installation through Cargo, and source builds with Cargo. The supplied package facts add Homebrew packaging, while the README points to Bioconda badges, tutorials, GitHub discussions, and downstream R/Bioconductor loading paths.

The adoption story is workflow-oriented: alevin-fry is meant to sit in pipelines with piscem or salmon upstream and fishpond, alevinQC, pyroe, or SingleCellExperiment downstream. The project also recommends simpleaf as a wrapper/workflow runner to make common reference-building and quantification workflows easier.

How it is used

Users typically produce RAD files with piscem or salmon alevin, then use alevin-fry commands such as permit-list generation, collation, quantification, and inference. The Read the Docs site organizes this around commands including `generate-permit-list`, `collate`, `quant`, `infer`, and `atac`.

Installation is intentionally package-manager friendly: `conda install -c bioconda alevin-fry`, `cargo install alevin-fry`, Homebrew packaging, or a direct source build with `cargo build --release`.

Why package nerds care

alevin-fry matters to package nerds because it is a scientific CLI whose real unit of use is the reproducible workflow. Versioned binaries, Bioconda, Cargo, Homebrew, and documentation all reduce friction for lab pipelines and HPC environments.

It also shows the Rust-in-bioinformatics pattern: performance-sensitive command-line genomics tools distributed through both language-native crates and scientific package channels.

Timeline

• 2021-06-29: bioRxiv preprint posted for alevin-fry.
• 2021-07-22: Changelog records 0.4.1 with metadata/output changes.
• 2021-10-16: 0.4.2 added USA mode support to `infer`.
• 2022-03-01: Nature Methods paper published.
• 2022-06-01: 0.6.0 added UMI resolution and command-line validation work.
• 2022-10-11: 0.8.0 fixed force-cells and expect-cells parsing.
• 2024-03-08: 0.9.0 released with libradicl compatibility work.

Related projects

• The official README names piscem and salmon as RAD-file producers; simpleaf as a wrapper/workflow runner; pyroe for enhanced transcriptome construction; fishpond for R ingestion; and alevinQC for quality control.
• The documentation also connects alevin-fry output to Bioconductor's SingleCellExperiment ecosystem.

Sources

• https://alevin-fry.readthedocs.io/en/latest/
• https://alevin-fry.readthedocs.io/en/latest/_sources/overview.rst.txt
• https://github.com/COMBINE-lab/alevin-fry/blob/master/CHANGELOG.md
• https://github.com/COMBINE-lab/alevin-fry/blob/master/README.md
• source_facts.package-manager