Software & Tools

[← Back to Home]

Open-source software and tools developed for cheminformatics and data analysis during my research.

ecfp_cli (`ecfp-gen`)

High-performance & Robust ECFP Fingerprint Generator CLI

• Key Features:
  • Intuitive CLI interface (`ecfp-gen --input ... --output ...`).
  • Fast parallel computation and streaming chunk execution for large-scale datasets.
  • Built-in desalting (`--desalt`) and optional standardization (`--standardize`).
  • Flexible column naming options (custom bit prefix and zero-indexed bit columns).
  • Multi-stage structured error logging (`.errors.tsv` / `.errors.csv`) for full workflow transparency.
• Track Record: Used for data preprocessing in the 146th Annual Meeting of the Pharmaceutical Society of Japan.
• Links: [GitHub Repository] ｜ [Latest Release (v0.2.0)]

mordred_descriptor_calculator (`mordred-desc`)

Reproducible Descriptor Calculator CLI with π-Conjugation Support

• Key Features:
  • Intuitive CLI interface (`mordred-desc --input ... --output ...`).
  • Reproducible 3D descriptors via fixed seeds and force field optimization (MMFF & UFF).
  • Custom algorithms for calculating π-conjugation features.
  • Multi-stage structured error logging (`.errors.csv`) for complete workflow transparency.
• Track Record: Used for feature extraction in the EUOS25 challenge (Winner, Fluorescence track).
• Links: [GitHub Repository] ｜ [Latest Release (v0.1.0)]

Installation

You can install and use these CLI tools by cloning the repository and running editable installation (Python 3.10 recommended):

git clone https://github.com/377H-Miru/[repository-name].git
cd [repository-name]
pip install -e .