biov 0.1.0

Next-generation development experience for computational molecular biology.

BioV

Next-generation development experience for computational molecular biology.

Highlights

• LLM friendly/native/driven: Designed for seamless integration with large language models, built for LLM workflows, and optimized for LLM-assisted development
• Pydantic-powered: Built-in validation and serialization for robust data handling
• Pandas ecosystem: Developer-friendly DataFrame operations with extended bioinformatics capabilities
• Modern tooling: Full type hints support and configuration through environment variables

Coordination system

[!IMPORTANT] BioV consistently uses BED-like coordinates with 0-based start positions and 1-based end positions, regardless of input format (including GFF3 and VCF).

This design decision was made to (by Gemini 2.5 Pro Exp):

1. Direct Compatibility: It aligns seamlessly with Python slicing and the indexing conventions of most relevant programming languages.
2. Reduced Errors: Minimizes the risk of off-by-one errors, which are notoriously common when converting between 1-based/inclusive and 0-based/semi-open systems.
3. Simplicity: Length calculation (end - start) and handling adjacent/empty intervals are mathematically cleaner and more intuitive within a programming context.
4. Developer Familiarity: Most developers working with sequences in code are already accustomed to this paradigm.

Environments

BioV can be configured through environment variables (prefixed with BIOV_) or a .env file:

• BIOV_HOME: Path to custom cache directory (default: platform-specific cache dir)
• BIOV_CACHE_HTTP: Enable/disable HTTP caching (default: True)

The cache directory is determined by:

1. BIOV_HOME if set
2. XDG_CACHE_HOME/biov if XDG_CACHE_HOME is set
3. Platform-specific cache directory otherwise

Supported formats

• GFF3
• BED
• VCF