An end-to-end automated pipeline for protein discovery that transitions from literature surveys to structural analysis and experimental prioritization.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install protein-mining
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install protein-mining using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Protein Mining Skill?
The Protein Mining Skill is a comprehensive bioinformatics workflow designed for researchers and computational biologists. It automates the systematic identification of target protein families by integrating literature mining, sequence collection, and advanced evolutionary modeling. By bridging the gap between raw genomic data and actionable experimental targets, this skill enables the generation of publication-ready figures and reports.
Leveraging state-of-the-art tools like AlphaFold2 and HMMER, the skill streamlines the discovery of novel proteins across species. It is specifically engineered to handle the heavy lifting of sequence alignment, phylogenetic reconstruction, and structural similarity searching, allowing scientists to focus on hypothesis generation and experimental validation within the Openclaw Skills ecosystem.
Protein Mining Skill Use Cases
- Discovering novel protein family members across diverse species genomes.
- Conducting systematic literature reviews and identifying research gaps in molecular biology.
- Building custom Hidden Markov Models (HMM) for specific protein domains.
- Predicting remote homologs using structure-based search tools like Foldseek.
- Prioritizing candidate proteins for laboratory synthesis and functional testing.
How Protein Mining Skill Works
- Target Identification: Conducts automated literature searches (PubMed/Google Scholar) to define seed sequences and research hypotheses.
- Sequence Mining: Collects, filters, and removes redundancy from initial sequence datasets via public databases.
- Model Building: Aligns sequences with MAFFT and constructs profile HMMs to capture conserved protein features.
- Genomic Search: Executes cross-species searches using the custom HMM to identify candidate proteins in genomic data.
- Structural Analysis: Predicts 3D structures using AlphaFold2 and performs fold-based similarity searches to find remote evolutionary links.
- Evolutionary Mapping: Builds high-quality phylogenetic trees using IQ-TREE to visualize protein divergence and functional niches.
- Scoring & Prioritization: Aggregates multi-dimensional features into a matrix to rank the top 20 candidates for experimental design.
Protein Mining Skill Setup
Ensure you have the required bioinformatics binaries installed on your system. You can then trigger the workflow through the Openclaw Skills interface.
# Install system dependencies
sudo apt-get install mafft hmmer iqtree
# Launch the full automated mining pipeline
protein-mining --target "ABC transporter" --species plants
# Execute specific modules individually
protein-mining --module 1 --target "ABC transporter"
Protein Mining Skill Data Schema & Taxonomy
The skill maintains a modular directory structure where each stage produces specific datasets and visualizations:
| Module | File Type | Description |
|---|---|---|
| 01_Literature | .pdf, .csv | Target lists and research hypotheses |
| 02_Sequences | .fasta | Raw, filtered, and non-redundant sequence sets |
| 03_Models | .hmm, .png | HMM profiles and ROC validation curves |
| 04_Search | .tbl, .csv | Species distribution heatmaps and hit tables |
| 05_Structure | .pdb, .m8 | AlphaFold2 predictions and Foldseek scores |
| 06_Phylogeny | .nwk, .pdf | Newick trees and publication-quality visualizations |
| 07_Ranking | .csv, .md | Feature matrices and experimental design proposals |
Protein Mining Skill Advanced Features
- Multi-agent literature synthesis for high-accuracy research gap identification.
- Automated ROC curve generation for HMM model validation.
- Integration with AlphaFold2 and Foldseek for structure-aware protein discovery.
- Positive selection site analysis to predict functional divergence.
- Generation of high-resolution phylogenetic trees (Newick format) compatible with professional publishing standards.
SKILL.md
Loading
Related Openclaw Skills
Baidu Hot Topics Monitor
A streamlined tool to monitor Baidu's real-time trending searches, rankings, and popularity indices directly from your AI agent workflow.
Optimacros AI Modeling & Formula Expert
A specialized AI assistant for Optimacros developers focusing on complex formulas, model architecture, and enterprise-grade access security.
Article to Feishu
A professional utility for converting web content from major platforms into structured Feishu documents while preserving image placement and bypassing anti-leech protections.
Business Writing
A professional business analyst skill designed to produce high-quality, data-driven research reports and consulting analyses with automated citations.
Task Automation
A comprehensive toolkit for automating repetitive work through scripts, scheduled jobs, and complex multi-step workflows.
Agent Debugger
A systematic debugging framework designed to identify and resolve AI agent failures, infinite loops, and resource bottlenecks.
Featured*
