dependency-resolution-and-validation

dependency-resolution-and-validation

Summary

Systematically identify, install, and validate transitive dependencies for bioinformatics pipelines, ensuring compatibility across Python, compiled tools, and R packages before executing the main workflow. This skill is critical for reproducible Hi-C data processing, where missing or mismatched dependencies (e.g., iced, numpy, scipy) silently break normalization steps.

When to use

Before running HiC-Pro or similar multi-stage pipelines on a new system or environment, especially when dependency installation is not automated (e.g., not in a conda environment or container). Use this skill when the article or README explicitly lists external dependencies with version constraints, and when downstream processing steps depend on optional modules that must be independently installed.

When NOT to use

• Dependencies are already installed in a conda environment or container (use the pre-built environment instead)
• The pipeline provides an automated dependency-resolution mechanism (e.g., setup.py with install_requires, or a conda environment.yml file) that you have not yet attempted to run
• The input is a pre-configured Docker or Singularity container image with dependencies baked in

Inputs

• Pipeline README or documentation specifying dependency list with version constraints
• config-install.txt or similar configuration template file
• Target system environment (Python version, package manager availability, PATH)
• Optional: existing environment.yml or requirements.txt file

Outputs

• Configured and validated Python environment with all required packages installed
• Updated config file (e.g., config-install.txt) with paths to dependencies and environment variables
• Installation log or validation report confirming version and API checks
• Ready-to-run pipeline environment for downstream workflow execution

How to apply

First, verify the availability and version of the Python interpreter (>3.7 for HiC-Pro) in your target environment. Next, systematically install each required dependency and its transitive dependencies in order of declaration, checking version constraints against the pipeline's documented requirements (e.g., numpy >=1.18.1, scipy >=1.4.1, iced from https://github.com/hiclib/iced). For optional dependencies that are no longer bundled with the main pipeline (e.g., iced, which HiC-Pro no longer includes in its source code), clone or download from the authoritative repository and install via pip or setup.py. After each installation, validate by importing the module in a Python interpreter and checking version/API availability. Document the installation paths and environment variables (PYTHONPATH, PREFIX, tool-specific paths) in a configuration file referenced by the main pipeline's install script (e.g., config-install.txt for HiC-Pro). Finally, run a smoke test that exercises the normalization or core functionality (e.g., import iced and call a key ICE method) to confirm dependencies are accessible to the pipeline.