lobster-research
Literature discovery and data acquisition agents for scientific research workflows.
Installation
pip install lobster-research
Agents
| Agent | Description |
|---|---|
research_agent |
Literature discovery specialist. PubMed/bioRxiv search, GEO/SRA dataset discovery, metadata extraction, publication queue management. |
data_expert_agent |
Data operations specialist. Queue-based downloads, modality management, local file loading, workspace orchestration. |
Services
| Service | Purpose |
|---|---|
| ModalityDetectionService | Auto-detect data modality type from file characteristics |
Features
Research Agent (Online Operations)
- PubMed literature search with filters and related paper discovery
- bioRxiv and medRxiv preprint search with full-text access
- GEO dataset discovery with organism and platform filtering
- SRA run metadata extraction and download URL generation
- PRIDE proteomics repository integration
- Full-text content extraction from PMC articles
- Methods section parsing for computational parameter discovery
- Publication queue for batch processing of research papers
- Automatic extraction of associated dataset identifiers
Data Expert Agent (Offline Operations)
- Execute downloads from pre-validated queue entries
- Zero online access boundary for security and reproducibility
- Multi-format file loading (CSV, TSV, H5AD, Excel)
- Modality listing, inspection, and validation
- Download strategy selection (AUTO, H5_FIRST, MATRIX_FIRST)
- Sample concatenation with union or intersection logic
- Failed download retry with exponential backoff
- Custom Python code execution for edge cases
Platform Support
- 10x Genomics MTX format (matrix, barcodes, features)
- H5AD pre-processed AnnData files
- Kallisto and Salmon bulk RNA-seq quantification
- CSV and TSV generic delimited matrices
- MaxQuant, Olink, and generic proteomics formats
Architecture
The research and data_expert agents implement a clean boundary pattern:
research_agent (ONLINE) data_expert (OFFLINE)
-- Search literature -- Execute downloads
-- Discover datasets -- Load local files
-- Extract metadata/URLs -- Manage modalities
-- Validate metadata -- Retry failed downloads
-- Create queue entries -- Concatenate samples
| |
----------- Queue Entry -------------
(PENDING -> IN_PROGRESS -> COMPLETED)
Requirements
- Python 3.12+
- lobster-ai >= 1.0.0
Documentation
Full documentation: docs.omics-os.com/docs/agents/research
License
MIT
1 maintainer
Metadata
1 maintainer
Release files for lobster-research 1.1.425
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| lobster_research-1.1.425.tar.gz | 69.8 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| lobster_research-1.1.425-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 143.5 kB
Release files / lobster_research-1.1.425.tar.gz
| Download URL | lobster_research-1.1.425.tar.gz |
|---|---|
| Size | 69.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
494d91d4defe833eae2dcb17734bfe7eba1cf7490de058aa85f580df806b55a9
|
|
BLAKE2b-256 checksum How to use checksums |
ebfaf2395e70099ef9e76e91066dea4db79b24f2836fdff9871b70e34c4de8f2
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 10, 2026.
Transparency logRelease files / lobster_research-1.1.425-py3-none-any.whl
| Download URL | lobster_research-1.1.425-py3-none-any.whl |
|---|---|
| Size | 73.7 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
a9fae36abb495b78e5726aa98bed5e53c5c5cfc01324548919987c73a489e6f6
|
|
BLAKE2b-256 checksum How to use checksums |
928a2d5ab8e1f26b2cc8c4df39e7e9de313f861fa6812b5d436a7d974d59c654
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 10, 2026.
Transparency log
