metaLncRNA v2.0.3 🧬🤖
metaLncRNA is a modular, high-performance Python framework designed to identify Long Non-coding RNAs (lncRNAs) by orchestrating an ensemble of seven diverse computational tools. It resolves the "reproducibility gap" by automating environment management and providing a robust consensus prediction through weighted soft-voting.
📂 Repository Structure
.
├── conda/ # Bioconda recipe and metadata
├── deploy/ # Containerization (Dockerfile, Singularity.def)
├── docs/ # Technical documentation and user guides
├── examples/ # Quick-start samples (FASTA, config templates)
├── galaxy/ # Galaxy Tool wrapper and test data
├── INPI_Registration/ # Legal software registration assets
├── paper/ # JOSS publication manuscript and bibliography
├── scripts/ # Bash scripts for HPC/Batch processing
├── src/
│ └── metalncrna/ # Main Python Package
│ ├── cli.py # Command-line interface entry point
│ ├── adapters/ # Wrappers for 7 lncRNA predictors
│ ├── engine/ # Core logic (Consensus, Dispatcher, Trainer)
│ ├── utils/ # AI Agent, Env management, Reports, FASTA handling
│ ├── data/ # Built-in weights and default configurations
│ └── third_party/ # Bundled legacy tools (CNCI, CPPred, LGC)
├── tests/ # Comprehensive Unit and Integration tests
├── pyproject.toml # Build system and dependency definitions
└── pixi.toml # Environment management configuration
🧩 Core Components Detail
src/metalncrna/adapters/: Orchestrates external tools like RNAsamba, CPAT, CPC2, etc., providing a unified interface for prediction.src/metalncrna/engine/:consensus.py: Implements the weighted soft-voting algorithm.dispatcher.py: Manages parallel execution of the ensemble.
src/metalncrna/utils/agent.py: Integrates with local LLMs (Ollama) for automated biological interpretation of results.galaxy/: AllowsmetaLncRNAto be integrated into Galaxy instances, supporting reproducible web-based workflows.
🔧 Recent Fixes (v2.0.3)
- 100% Original Sequence ID Preservation: Implemented
build_id_mappingandmap_df_sequence_idsto ensure original transcript names from transcriptome assemblers (Trinity, StringTie, Cufflinks, etc.) remain intact throughout the entire consensus and reporting pipeline. - Deduplication & Merge Safety: Prevents Cartesian product row explosions by deduplicating predictions per transcript before cross-tool voting.
- Adapter Attribute Bugfix: Fixed
AttributeErrorcaused by missingself.tool_pathinBaseAdapter. - Automatic HTML Dashboard: Integrated HTML report generation directly into
metalncrna predict. - Enhanced AI Agent Feedback: Clear installation guidance and connection diagnostics for Ollama (
pip install metalncrna[agent],ollama serve,ollama pull). - PyPI Release Workflow: Enabled
skip-existing: truein GitHub Actions publishing pipeline.
🔧 Previous Fixes (v2.0.0)
- CNCI Stability: Fixed a critical hang in the CNCI legacy tool caused by non-canonical nucleotides (e.g., K, V, M) and a multiprocessing deadlock in the original Python 2.7 implementation.
- Improved Filtering: Implemented rigorous FASTA validation in the CNCI adapter to exclude sequences with ambiguous characters, preventing
KeyErrorcrashes. - Resource Optimization: Optimized tool dispatching by limiting CNCI threads to 4, reducing I/O overhead and improving performance for small-to-medium files.
- Debug Resiliency: The dispatcher now preserves intermediate files automatically if a tool failure occurs, facilitating troubleshooting.
🔧 Previous Fixes (v1.2.1)
- Consensus Logic: Updated
consensus_supportto reflect the number of tools that agree with the final consensus label, providing better interpretability. - CPC2 Integration: Fixed a critical parsing error where coding probability and label columns were mismatched (v1.1.8).
- Cleanup: Removed unimplemented/experimental adapters to ensure stability.
⚙️ Configuration
metaLncRNA follows a robust configuration loading order:
- Internal Defaults: Built-in weights and paths in
src/metalncrna/data/default_config.yaml. - Local Config:
metaLncRNA_config.yamlin your current working directory. - User Home:
~/.metalncrna/config.yaml. - Explicit Path: Provided via the
-cor--configflag.
🚀 Key Features
- Ensemble Prediction: Combines 7 tools (RNAsamba, CPAT, CPC2, PLEK, CNCI, CPPred, LGC).
- Interactive AI Agent: Integrated local LLM assistant (Llama-3.2 or OpenBioLLM) to interpret results and explain classification decisions.
- Reproducibility First: Built-in environment isolation via Mamba and Pixi.
- Standardized Reports: Comprehensive TSV reports with tool congruence metrics.
- Publication Ready: Formatted according to JOSS standards for scientific software.
Metadata
Release files for metalncrna 2.0.6
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| metalncrna-2.0.6.tar.gz | 1.9 MB | Without provenance | Details for metalncrna-2.0.6.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| metalncrna-2.0.6-py3-none-any.whl | Python 3 | none | any | Details for metalncrna-2.0.6-py3-none-any.whl |
Total release size: 3.8 MB
Release files / metalncrna-2.0.6.tar.gz
| Download URL | metalncrna-2.0.6.tar.gz |
|---|---|
| Size | 1.9 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
6c49969dbd23e63c74a616f6dc46683fdcea1c7d064b5aa0f7e3e86ded3dcc04
|
|
BLAKE2b-256 checksum How to use checksums |
2df96f1f82092768d26769e004387a311815c70b3202c1df30b31f9881f25d6c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Release files / metalncrna-2.0.6-py3-none-any.whl
| Download URL | metalncrna-2.0.6-py3-none-any.whl |
|---|---|
| Size | 1.9 MB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
9b9089b8fa557c636d6d46642479d9dab2dc632455925eb3882a06b3a74e7a42
|
|
BLAKE2b-256 checksum How to use checksums |
22d44362e10a1d4a44f3c7ba7c3040992d1cbd3ce6c5eb6ec8d9c2a95e2a42c3
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Security for metalncrna 2.0.6
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.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.