Liatir
Liatir is an independent, open-source, local-first bioinformatics desktop application. It brings bioinformatics tools, locally managed AI models, plugins, visual pipelines, viewers and external workflows together on the user's own computer, including offline and air-gapped use, so scientific data stays under the user's control.
At a Glance
- Bioinformaticians and computational biologists
- Academic and scientific research teams
- Users handling patient, proprietary or otherwise sensitive biological data
- Researchers needing reproducible offline or air-gapped analysis
- +2 more
AI Tools by Liatir
(1)
Liatir
Local First Bioinformatics Desktop App
Discussions
No discussions yet
Be the first to start a discussion about Liatir
Latest News
Liatir app repository continued active development, including a redesigned DNA-canvas documentation hero and homepage integration updates.
Liatir published a Scientific Showcase comparing single-cell foundation models Geneformer and scGPT with PCA, Harmony and scVI.
Liatir 0.2.1 released, fixing AI Model/runtime installation and Windows WSL2 NVIDIA availability; downloads were provided for macOS, Linux and Microsoft Store Windows.
Liatir 0.1.1 released with fixes for 3D structure, molecular trajectory, genome-track and structure-prediction viewers.
Products & Services
Free, open-source local-first desktop environment for bioinformatics tools, AI models, plugins, pipelines, viewers and results/provenance management. Distributed for Apple-silicon macOS, Windows 10/11 through Microsoft Store, and Linux x86_64 as AppImage, deb and rpm packages.
Public developer-facing packages for building, validating and packaging Liatir plugins.
Self-contained extensions that appear as ordinary tools and pipeline nodes; they can run with Node, Python or WebAssembly.
Signed, locally installed Runtime Boxes and tools for single-cell embeddings, structure prediction and protein-ligand affinity, including Geneformer, scGPT, UCE, Boltz-2 and Protenix.
Market Position
Liatir positions itself as a free, local-first alternative to cloud-based bioinformatics platforms: rather than uploading data and paying for remote compute, it runs tools and AI models on the user's hardware, works offline, avoids subscriptions and rate limits, and retains run history and provenance locally. Its differentiators are the integrated desktop UI, verified one-click runtime/model installation, extensible plugin and pipeline system, and privacy/air-gap posture.
Leadership
Executive Team
Lorenzo Suffritti
Developer and publisher
Identified by the Microsoft Store as Liatir's developer and publisher; the Liatir GitHub repositories identify the associated contributor as suffro/Lorenzo.
Founding Story
Liatir was started as an independent project to make science more accessible while avoiding the privacy, connectivity, subscription and reproducibility trade-offs of cloud bioinformatics. Its initial vision was a free desktop environment in which tools, AI models and workflows install and run locally, with data, logs, results and provenance retained on the user's machine.
Business Model
Revenue Model
The application is free and open source, with no subscriptions or rate limits. Liatir solicits voluntary user support through Ko-fi donations to keep the project independent.
Pricing Tiers
100% free forever; the desktop application is available for macOS, Windows and Linux.
Target Markets
- Bioinformaticians and computational biologists
- Academic and scientific research teams
- Users handling patient, proprietary or otherwise sensitive biological data
- Researchers needing reproducible offline or air-gapped analysis
- Developers building custom scientific tools and plugins
- Teams using local Nextflow workflows and single-cell or molecular-structure analysis
- Offline and privacy-sensitive genomic analysis
- FASTQ quality control and read trimming
- Sequence alignment and mapping with BWA-MEM or minimap2
- BAM, SAM, CRAM, VCF and BCF inspection, statistics and filtering
- Single-cell indexing, quantification, embeddings and visualization
- Local AI-assisted structure prediction and protein-ligand affinity analysis