pgdh-evaluate

Synchronize and rank protein binder designs for the 15-PGDH target.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill streamlines the process of collecting, ranking, and evaluating protein binder designs for the 15-PGDH target, automating complex computational biology workflows.

Core Features & Use Cases

  • Design Synchronization: Collects and standardizes design outputs from various tools into a central 'source of truth'.
  • Automated Evaluation: Submits jobs for refolding, cross-validation, and scoring using computational resources.
  • Ranking: Computes composite scores to rank designs based on multiple metrics.
  • Use Case: After generating new protein binder designs, use this Skill to automatically sync them, assess their designability through refolding, validate their binding confidence with Boltz-2, and score their interaction strength with ipSAE, ultimately ranking them for further consideration.

Quick Start

Use the pgdh-evaluate skill to collect and rank designs from S3.

Frequently Asked Questions about pgdh-evaluate

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

FAQPage Schema
How do I rank and evaluate protein binder designs for the 15-PGDH target?

To evaluate protein binder designs for 15-PGDH, this Skill synchronizes outputs from various design tools, submits computational refolding and validation jobs, and ranks designs using composite metrics based on designability, Boltz-2 binding confidence, and ipSAE interaction strength.

What is the best way to automate protein design assessment and scoring using Boltz-2 and ipSAE?

Automated protein design assessment is achieved by submitting computational jobs to Lyceum for refolding and validation, scoring binding confidence with Boltz-2, evaluating interaction strength with ipSAE, and computing composite scores to rank the designs.

Can I use this Skill to sync and standardize protein binder outputs from multiple design tools?

Yes, this Skill synchronizes and standardizes protein binder outputs from multiple design tools by collecting them into a central source of truth, requiring integration with S3 for data storage and Lyceum for GPU job submission.

Do I need Lyceum and S3 integration to run automated computational drug discovery workflows?

Yes, you need Lyceum integration for GPU job submission and S3 for data storage to run these automated computational drug discovery workflows, enabling cross-validation and scoring of protein binder designs.

How does composite scoring work when ranking computational protein designs?

Composite scoring for ranking computational protein designs works by aggregating multiple evaluation metrics, including refolding designability, Boltz-2 binding confidence, and ipSAE interaction strength, into a unified score for prioritization.