by-failure-diagnosis

Community

Statistical failure post-mortem for designs.

Author001TMF
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Diagnose failure modes by testing whether PASS designs differ from FAIL designs on key numeric features using Mann-Whitney U tests.

Core Features & Use Cases

  • Runs Mann-Whitney U tests between PASS and FAIL groups for canonical metrics (ipsae, ipsae_min, iptm, plddt, rmsd, liabilities, cdr3_length, hydrophobic_fraction, net_charge) with Benjamini-Hochberg correction.
  • Produces a structured diagnosis containing discriminating features, effect sizes, and q-values, plus a human-readable table and top recommendations.
  • Provides CLI scripts (diagnose_from_csv.py, plot_distributions.py) to generate dashboards, plots, and exportable results for downstream optimization.

Quick Start

Run the diagnosis tool on your campaign's screening output to obtain a feature-discrimination report and recommended next actions.

Dependency Matrix

Required Modules

scipynumpypandasmatplotlib

Components

scriptsreferences

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: by-failure-diagnosis
Download link: https://github.com/001TMF/blatant-why/archive/main.zip#by-failure-diagnosis

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 510,000+ vetted skills library on demand.