lyceum_ipsae

Score predicted protein-protein interactions using ipSAE, pDockQ, pDockQ2, and LIS metrics.

Updated Feb 27, 2026
One-click install
npx skills add https://github.com/alex-hh/in-silico-pgdh --skill lyceum-ipsae
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: lyceum_ipsae
Source: https://github.com/alex-hh/in-silico-pgdh/tree/main/.claude/skills/lyceum_ipsae
Command: npx skills add https://github.com/alex-hh/in-silico-pgdh --skill lyceum-ipsae

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires lyceum_ipsae.py, ipsae.txt, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the scoring of predicted protein-protein interactions, providing crucial metrics for evaluating binding confidence.

Core Features & Use Cases

  • Interaction Scoring: Computes ipSAE, pDockQ, pDockQ2, and LIS scores for protein complexes.
  • Input Flexibility: Accepts PAE and structure files from various predictors like AlphaFold2/3 and Boltz.
  • Use Case: After designing a potential protein binder, use this Skill to score its predicted interaction with the target protein, helping to rank and select the most promising candidates.

Quick Start

Use the lyceum_ipsae skill to score the protein interaction using the provided PAE file 'pae.json' and structure file 'model.cif'.

Frequently Asked Questions about lyceum_ipsae

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I score predicted protein-protein interactions from AlphaFold or Boltz outputs?

To score predicted protein-protein interactions, you can use this Skill to compute ipSAE, pDockQ, pDockQ2, and LIS metrics. It accepts PAE and structure files from predictors like AlphaFold and Boltz to evaluate binding confidence.

What metrics are used to evaluate protein binding confidence from structure predictions?

Protein binding confidence is evaluated using ipSAE, pDockQ, pDockQ2, and LIS scores. These metrics are computed from your provided PAE and structure files to rank and select promising protein binder candidates.

Do I need PAE and structure files to calculate ipSAE and pDockQ scores?

Yes, you need PAE and structure files to calculate ipSAE and pDockQ scores. The Skill requires these inputs from structure prediction outputs like AlphaFold or Boltz to perform the binding confidence analysis.

Can I use this Skill to rank and select designed protein binders?

Yes, you can use this Skill to rank and select designed protein binders. After designing a potential binder, it scores the predicted interaction with the target protein to help identify the most promising candidates.

Does this protein interaction scoring tool work with AlphaFold3 and Boltz prediction outputs?

Yes, this protein interaction scoring tool works with AlphaFold3 and Boltz prediction outputs. It offers input flexibility by accepting PAE and structure files from various structure predictors for comprehensive binding confidence evaluation.