Access request submitted ✓
Private beta — now accepting researchers

AI native platform for multi-omics research

OmniBioAI unifies reproducible multi-omics analysis, agentic AI reasoning, and enterprise-grade infrastructure — across local, HPC, and cloud environments.

↗ Request Access ▶ Watch Demo 🌐 Try Web App
QR code for OmniBioAI Web Studio Open on your phone Scan to launch Web Studio
12,000+TES tool definitions
1000+Workflows
500+Workbench plugins
1,320+Container baseline
40+Agentic Pipelines/Plugins
28Microservices
28M+Abstracts
150+Domains
12Organisms
Live platform metrics & health →
// Biomedical AI — Literature Intelligence

Powered by PubMed.
Enhanced by AI.

Grounded in Evidence. A 9-stage retrieval-augmented generation pipeline over 28M+ PubMed abstracts — delivering cited, reproducible answers grounded in published science.

01
User Query Natural language or structured biomedical question input
02
Query Expansion (LLM) Synonyms, MeSH terms, and related concepts added via LLM reformulation
03
Hybrid Retrieval (FAISS + BM25) Dense vector search (PubMedBERT embeddings) merged with sparse BM25 keyword retrieval
04
Reciprocal Rank Fusion (RRF) Dense and sparse result lists merged and re-scored by RRF for unified ranking
05
Cross-Encoder Re-ranking PubMedBERT cross-encoder re-scores top-k candidates for precision retrieval
06
Citation Confidence Scoring Papers weighted by citation impact, journal tier, recency, and domain relevance
07
Knowledge Graph Enrichment (Neo4j) Entity links — genes, diseases, drugs, pathways — resolved via Neo4j biomedical KG traversal
08
LLM Reasoning (DeepSeek-R1) Evidence chains synthesized with explicit chain-of-thought reasoning and full citation anchoring
09
Structured Output (PMIDs + Evidence + Entities) PMIDs + evidence snippets + named entities + confidence scores returned as structured JSON
28M PubMed abstracts indexed and searchable
150+ Biomedical domains covered
<30s Query response time end-to-end
9-stage Pipeline with full citation provenance
PubMedBERT embeddings Neo4j knowledge graph Hybrid BM25 + FAISS DeepSeek-R1 reasoning Cross-encoder re-ranking Reciprocal Rank Fusion Citation confidence scoring MeSH query expansion Structured JSON output PMID-anchored evidence Named entity extraction Local · air-gapped · private
// Core capabilities

Built for real science,
not demos

Every component is designed with clinical-grade reliability and reproducibility at its core.

⬡
Agentic AI reasoning

LangGraph-based orchestration with RAG-powered assistants using Hugging Face and Ollama — bridging deterministic bio-computation with explainable AI interpretation.

◈
Workflow agnostic

Native support for WDL, Nextflow, Snakemake, and CWL. Cloud-agnostic Tool Execution Service ensures 1:1 parity between local and massive-scale genomic execution.

◎
GPU-accelerated stack

Optimized on NVIDIA PyTorch with CUDA on DGX Spark. High-performance GNNs for drug discovery and deep learning for single-cell transcriptomics.

⬕
End-to-end provenance

Production-grade Model Registry and LIMS-X metadata system for total traceability — from raw FASTQ to drug-target intelligence and pathway enrichment.

Lab Integrations:
· Benchling (samples, sequences, entities, notebooks)
· REDCap (patient cohorts, clinical research data)
· eLabFTW (experiment export, file upload, tags)
· AWS S3 (import datasets, export results, MinIO compatible)
· Zenodo (publish with DOI, open science, reproducible research)
· LIMS built-in
· ELN-ready export: PDF reports (WeasyPrint) · ELN JSON (machine-readable) · Benchling notebook export · eLabFTW experiment export · CSV/TSV universal format

🔐
SSO Support

Sign in with Google, GitHub, or Microsoft, alongside secure email/password authentication via JWT + zero-trust auth service. Existing accounts can link a provider identity with a one-time confirmation step.

◇
Reliability-first design

97% average test coverage across 28 microservices. Sub-3ms latency for critical service handshakes. Built for clinical-grade stability from day one.

⬡
Plugin ecosystem

Extensible plugin architecture with 500+ Workbench plugins and an OnboardAI documentation browser. Documentation and plugin inventories vary by repository release.

⚙
Modular pipeline system

Each bioinformatics workflow is modular, allowing researchers to customize, extend, or replace analysis steps without breaking reproducibility.

☁
Cloud & HPC ready

Runs consistently across local machines, Slurm-based HPC clusters, and cloud platforms like AWS, Azure, and GCP with identical execution logic.

🐳
Docker Images

*Historical beta-release inventory; live availability varies by deployment.
320 Docker images + 1,000 ARM64 SIF images in that baseline
Available: ghcr.io/omnibioai & huggingface.co/omnibioai
ARM64 + x86_64 support

🤖
Agentic workflow orchestration

LangGraph-based agentic workflows dynamically plan, execute, and adapt bioinformatics pipelines, enabling multi-step reasoning across tools, datasets, and analysis stages.

🛡
Production-Grade Observability

Full-stack error tracking and performance monitoring via Sentry.io across all 28 microservices — with release tracking, error alerting, and performance tracing built in from day one.

⟐
Real-Time Collaboration

WebSocket-based shared workspaces with live chat, object-level comments, annotation history, and full audit trails for reproducible science.

🎓
Zenodo Integration

Publish analysis results to Zenodo with automatic DOI generation for citable, reproducible research. Search and import public datasets directly from Zenodo.

📓
eLabFTW Integration

Export analysis results directly to eLabFTW electronic lab notebook. Creates experiment entries, uploads output files, and adds tags automatically. Open-source ELN used by thousands of academic labs worldwide.

// Enterprise & scientific workflow integrations

Scientific work,
connected across your stack

OmniBioAI's agentic workflows connect research, data, compute, MLOps, cloud infrastructure, and enterprise systems — so scientists and AI agents can work across the tools your organization already uses, without leaving the platform.

67 integrations across 9 categories Browse the full catalog in the docs ↗

Genomics & Bio Platforms

9

Sequencing, cloud-genomics and bioinformatics platforms, plus GA4GH interoperability.

AWS HealthOmicsSequence stores, workflows and policy-gated workflow runs Illumina BaseSpaceSequencing projects, runs, samples, files and analyses Illumina Connected AnalyticsICA projects, data, pipelines and analyses DNAnexusGenomics cloud projects, files, workflows and executions TerraWorkspaces, WDL workflows, data tables and outputs Seven BridgesProjects, files, workflows, tasks and provenance Seqera PlatformNextflow Tower pipeline discovery and run monitoring GalaxyWorkflows, histories, datasets, tools and jobs GA4GH InteroperabilityRefget and DRS access plus VRS allele construction

Workflow, Orchestration & Compute

10

Workflow engines, pipeline orchestrators, cluster schedulers and notebook environments.

Apache AirflowDAG discovery, triggering, monitoring and task logs Argo WorkflowsWorkflow discovery, metadata and manifest submission CromwellWDL/CWL workflow submission and monitoring DagsterRepositories, runs, assets and health PrefectDeployments, flows, runs and work pools MetaflowFlows, runs, steps and artifacts via the metadata service KubernetesClusters, workloads, jobs and monitoring SlurmHPC cluster, partition, node and job monitoring KNIMEWorkflow discovery, execution, results and provenance JupyterHubInstitutional notebook servers, users and sessions

Cloud Storage & Data Transfer

11

Object storage, file sharing, private cloud and bulk research-data transfer.

AWS S3Import datasets and export results to S3 buckets Azure Blob StorageContainer and blob discovery and metadata Google Cloud StorageBucket and object discovery and metadata MinIOSelf-hosted S3-compatible buckets and objects CephCluster health, pools, block images and filesystems OpenStackProjects, compute, volumes and identity BoxSearch, transfer and organize research datasets Google DriveResearch-file discovery, transfer and Shared Drives NextcloudSelf-hosted research-file access and collaboration GlobusEndpoint discovery and large research-data transfers Cloud IntegrationUnified provider → service → action cloud access

Data Warehouses & Analytics

4

Warehouse and lakehouse platforms and the tooling that models data inside them.

DatabricksUnity Catalog discovery and governed SQL access SnowflakeDatabase, schema and table discovery and queries Google BigQueryProjects, datasets, tables and job metadata dbtModels, sources, tests, jobs and lineage

ML & Experiment Tracking

4

Experiment tracking, model hubs and data/model versioning.

MLflowExperiments, runs, metrics and the model registry Weights & BiasesProjects, runs, metrics and artifacts Hugging FaceModels, datasets, repositories and inference DVCVersioned datasets, pipelines and experiments

LIMS, ELN & Biobank

7

Laboratory information management, electronic lab notebooks and biospecimen management.

BenchlingImport samples and sequences, export results eLabFTWExport results to the electronic lab notebook LabKeyPermission-aware discovery and controlled queries LIMS-XExternal LIMS sample and project metadata Clarity LIMSSamples, projects, containers and processes OpenSpecimenBiobank participants, specimens and visits REDCapClinical research cohorts in, analysis results out

Data Repositories & Metadata

4

Research data repositories and data-catalog / metadata platforms.

ZenodoPublish results with a citable DOI FigshareResearch data repository access and publishing SynapseSage Bionetworks projects, files and tables OpenMetadataData catalog entities, lineage and metadata

CI/CD, Source Control & Containers

10

Git hosting, build and pipeline systems, and container registries.

GitHub + GitLabRepositories, code search, issues and pull requests GitHub ActionsWorkflow runs, jobs and CI/CD monitoring GitLabProjects, pipelines and CI/CD monitoring JenkinsJobs, parameterized builds and console output Azure DevOpsRepos, work items, pipelines and builds Docker HubRepositories, tags, manifests and digests GitHub Container RegistryImage tags and manifests on ghcr.io Quay.ioRepositories, OCI manifests and multi-platform checks HarborProjects, repositories and replication NVIDIA NGCCatalog and artifact namespace discovery

Collaboration, ITSM & Operations

8

Messaging, Microsoft 365, issue and incident management, monitoring and secrets health.

SlackChannels and users for agent collaboration Microsoft TeamsChannel notifications via incoming webhooks Microsoft GraphEntra ID, SharePoint, Outlook and Teams Jira + ConfluenceSearch, create and update issues and pages ServiceNowSearch, create and update incidents PagerDutyIncidents, services and on-call monitoring Prometheus & GrafanaMetrics queries, alerts and dashboards HashiCorp VaultHealth, mounts and seal status
// Catalogs

What's in the platform

All catalogs in the docs ↗
1000+WorkflowsIn 100+ domainsEngine mix in docs catalog: Nextflow 861 · WDL 5 · Snakemake 2 · CWL 1 501Workbench pluginsAcross 22 categories, including 66 integrations 12,276TES tool definitions8 execution backends · 9,061 auto-imported and not yet verified 130Reference database plugins90 implemented · 40 scaffolded 219API routesFound in source across 8 services · 109 documented publicly

Plugin, tool, reference database and API counts are read from the documentation's generated catalogs (19 Sep 2026). They describe what is registered or configured in source, not that every entry has been tested or deployed. How to read these figures ↗

// Multi-omics coverage

Every modality,
one platform

From raw sequencing reads to biological insight — 1000+ workflows across 100+ research domains, grouped here into the areas they serve.

Genomics
Genome & variant analysis

Germline variant calling, annotation and prioritisation, from exomes to assemblies.

WGSWESSV callingML variant prioritisationGenome assemblyPangenomeOptical mappingMitochondrialTelomere
Long-read
Long-read sequencing

Nanopore and PacBio alignment, assembly, methylation and isoform analysis.

NanoporePacBioNative methylationIsoforms
Transcriptomics
Bulk & specialised RNA

Expression, splicing and RNA biology beyond standard RNA-seq.

RNA-seqLong-read RNAmiRNAlncRNAcircRNARibo-seqRNA editing
Single-cell
Single-cell & multiome

QC, integration, clustering, trajectories and multimodal single-cell assays.

scRNA-seqCell RangerCITE-seqscATAC-seqMultiome / ArchR
Spatial
Spatial omics

Spatially resolved expression and protein maps with cell-type deconvolution.

Visium / HDXeniumCODEX / PhenoCyclerSpatial proteomicsSpatial multi-omics
Epigenomics
Functional genomics & epigenomics

Regulatory elements, chromatin state, 3D genome and CRISPR screens.

ATAC-seqChIP-seqMethylationHi-CCRISPR screensBase editing
Proteomics · Metabolomics
Proteomics & metabolomics

Mass-spec quantification, enrichment and metabolite profiling.

Mass specProteogenomicsMetabolomicsExposomeGSEA / GSVA
Multi-omics
Multi-omics integration

Joint analysis across genomic, transcriptomic, epigenomic and proteomic layers.

Multimodal integrationMulti-omics QCCross-layer methods
Oncology
Cancer genomics

Somatic mutations, tumour burden and liquid biopsy analysis.

Somatic callingTMB / MSIctDNAFragmentomics
Immunology
Immunology & infectious disease

Immune profiling, HLA and repertoire analysis, pathogens and vaccines.

HLA typingVDJ repertoireImmune deconvolutionVaccine designViral genomics
Microbiome
Microbiome & metagenomics

Taxonomic and functional profiling, including host–microbe interactions.

Shotgun metagenomicsTaxonomic profilingHost–microbiome
Clinical
Clinical & translational

Cohort analysis, clinical reporting and precision-medicine workflows, with REDCap import and export.

Clinical reportingPharmacogenomicsPolygenic riskClinical NLPTrial matchingTransplantReproductive
Drug discovery
Drug discovery

GNN-based target identification, drug response and repurposing.

Target identificationDrug repurposingDrug synergyResponse predictionADMET
Population
Population & specialised genomics

Population, evolutionary and ancient genomes, plus neuro and aging studies.

Population geneticsEvolutionaryAncient DNAForensicsNeurogenomicsAgingEmerging methods
AI · Platform
AI models & platform methods

Foundation models, knowledge graphs and the tooling that keeps pipelines reproducible.

scGPT / foundation modelsBiomedical knowledge graphBenchmarkingSimulationPipeline chainingReference data

Browse every workflow domain in the docs ↗

PythonRNextflow WDLSnakemakeCWL GATKSeuratDESeq2 LangGraphPyTorchCUDA DockerKubernetesFastAPI DjangoCeleryRedis MySQLAWSAzureGCP SlurmHuggingFaceOllamaSentry BenchlingREDCapeLabFTW AWS S3Zenodo
// Reference genomes

Reference Genomes

12 species with genome sequence and gene annotation on disk, and transcript sequences for 11 of them. Prepared aligner indexes are ready for human (STAR, Cell Ranger) and mouse (Bowtie2, Cell Ranger).

Species Common Name Assemblies Annotation Use Cases On disk
Homo sapiens Human GRCh38 (hg38), hg19 GENCODE v44, v46 WGS, WES, scRNA, ATAC, clinical GenomeTranscriptsGTFSTARCell Ranger
Mus musculus Mouse GRCm39, mm10 GENCODE vM33 Drug discovery, knockout models GenomeTranscriptsGTFBowtie2Cell Ranger
Rattus norvegicus Rat GRCr8 Ensembl 111 Pharmacology, toxicology GenomeGTF
Pan troglodytes Chimpanzee Pan_tro_3.0 Ensembl 112 Comparative genomics, evolution, primate research GenomeTranscriptsGTF
Macaca mulatta Rhesus Macaque Mmul_10 Ensembl 112 Primate translational research GenomeTranscriptsGTF
Danio rerio Zebrafish GRCz11 Ensembl 112 Developmental biology, screens GenomeTranscriptsGTF
Drosophila melanogaster Fruit Fly BDGP6 Ensembl 112 Genetics, CRISPR screens GenomeTranscriptsGTF
Saccharomyces cerevisiae Yeast R64-1-1 Ensembl 112 Yeast two-hybrid, metabolomics GenomeTranscriptsGTF
Caenorhabditis elegans C. elegans WBcel235 Ensembl 112 Aging, longevity, RNAi screens, neuroscience GenomeTranscriptsGTF
Arabidopsis thaliana Thale Cress TAIR10 Ensembl Plants 59 Plant biology, epigenomics, stress response GenomeTranscriptsGTF
Sus scrofa Pig Sscrofa11.1 Ensembl 112 Xenotransplantation, cardiovascular, metabolic disease GenomeTranscriptsGTF
Gallus gallus Chicken GRCg7b Ensembl 112 Immunology, developmental biology, vaccine research GenomeTranscriptsGTF
STAR Prepared index, ready to use GTF Sequence or annotation file

Inventory of local reference storage, measured 19 Sep 2026. Live status on the platform dashboard ↗ · Full inventory in the docs ↗

// System requirements

What you need
to get started

OmniBioAI Studio runs on any modern Linux or macOS machine with Docker installed. One download, no command line needed.

💾
Memory

Minimum: 16 GB RAM
Recommended: 32 GB RAM
With local LLM: 64 GB RAM

💿
Storage

Installer: ~75–105 MB
Docker images: ~10 GB (one-time pull)
Data + work dirs: 50–200 GB

🖥️
Operating System

Linux: Ubuntu 20.04+ (AppImage)
macOS: 12+ Apple Silicon + Intel
Windows: WSL2 + Docker

🐳
Docker

Docker Engine 24+ or Docker Desktop
Docker Compose v2 (included)
No other dependencies required

⚡
GPU (Optional)

NVIDIA GPU + nvidia-container-toolkit
Required only for local LLM inference
Cloud API (Claude/GPT) works without

🌐
Network

Internet required for first boot only
~10 GB pulled from ghcr.io automatically
Fully offline after first run

🔑
License

30-day free trial included
License key with every download
7-day offline grace period
[email protected]

🧬
jq

Required: sudo apt install jq
Or: brew install jq
Used for health checks and config
Included in most Linux distros

● Private Beta · v0.7.1-beta

Access requires approval

Approved researchers receive a platform-specific download link and onboarding support within 1–2 business days.

↗ Request Beta Access

License key included · 30-day free trial

// Downloads

OmniBioAI Studio v0.7.1-beta

Choose your platform and architecture. All installers include a 30-day free trial license.

macOSApple Silicon or Intel
↓ Apple Silicon · DMG 102 MB ↓ Intel · DMG 105 MB
🐧Linux x86_64Intel / AMD · Ubuntu 20.04+
↓ AppImage 90 MB ↓ DEB ~91 MB ↓ RPM ~81 MB
🦾Linux ARM64DGX · Graviton · Raspberry Pi
↓ AppImage ~85 MB ↓ DEB ~86 MB ↓ RPM ~76 MB
🪟WindowsWindows 10 / 11 x64 · WSL2
↓ Installer · EXE ~104 MB
Linux install commands
sudo dpkg -i omnibioai-studio_*amd64.deb sudo rpm -i omnibioai-studio-*.x86_64.rpm sudo dpkg -i omnibioai-studio_*arm64.deb sudo rpm -i omnibioai-studio-*.aarch64.rpm chmod +x *.AppImage && ./OmniBioAI*.AppImage
All releases on GitHub ↗
// System architecture

Clinical-grade design

A containerized, microservices-led environment built for massive scale, traceability, and high performance.

Engineered for absolute provenance

OmniBioAI Studio separates user experience orchestration from heavy-lifting workflow engines. It executes multi-omics routines natively, passing biological insights directly into automated reporting and visualization pipelines.

With strict isolation across its core service layers, researchers can safely deploy pipelines locally or easily burst to Slurm or cloud systems without code modifications.

↗ See live architecture & health
5M+ Lines of code
97% Avg test coverage
33 Repos · all clean
65K+ Automated tests
OmniBioAI Studio UI (Desktop Frontend)
│ Handshake (sub-3ms latency)
Agentic AI Orchestration (LangGraph / Ollama / HF)
│ Orchestration & data mapping
BioFlow Runtime Engine (Nextflow / WDL / Snakemake)
│ Tracking & provenance link
LIMS-X Metadata & Sample Tracking System
│ Infrastructure layer
GPU Accelerated Stack (CUDA / NVIDIA DGX Spark)
// Academic presence

Peer-reviewed research

Methods developed and powered by OmniBioAI platform architectures across transcriptomics and proteomics.

2025
Genetic mutations in lymphocytic variant of hypereosinophilic syndrome: study of five siblings
Frontiers in Medicine · December 2025
↗ View paper
2018
Whole Exome Sequencing identifies common and rare variant Metabolic QTLs in a Middle Eastern Population
Nature Communications · January 2018
↗ View paper
2015
MetaRNA-Seq: An Interactive Tool to Browse and Annotate Metadata from RNA-Seq Studies
BioMed Research International · August 2015
↗ View paper
// Apply for private beta

Request Access

Get access to OmniBioAI Studio v0.7.1-beta. Accelerate your multi-omics data integration with robust, explainable AI workflows.

Your Operating System *