Skip to main content
skill CX Codex

glycoengineering — Claude Code skill: protein glycosylation analysis

A skill for working through N- and O-glycosylation sites in protein sequences: motif scanning, mutation planning and predictor prep.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add K-Dense-AI/scientific-agent-skills --skill glycoengineering -a claude-code
Open on GitHub

What glycoengineering does

The skill walks through glycosylation analysis for eukaryotic secretory-pathway proteins. It starts by recording the accession, isoform, exact sequence, expression host and signal peptide topology. Canonical N-X-[S/T] candidates are then scanned with X not proline, and overlapping hits are kept: in NNST a candidate begins at position 1 and another at position 2. S/T-rich stretches get a separate pass for mucin-type O-GalNAc, though density there is a descriptive feature only, not a probability of glycosylation.

A local helper ships as scripts/glycoengineering_tools.py and uses the Python standard library. It takes sequences of the 20 canonical amino acids, normalizes case and whitespace, and rejects FASTA headers, gaps and unknown residues. The functions find_n_glycosylation_sequons, eliminate_glycosite, add_glycosite and find_st_rich_sites report 1-based coordinates against the sequence you submitted. Put the scripts/ directory on PYTHONPATH, or run Python from inside it.

Predictions go through the official web forms for NetNGlyc 1.0 and NetOGlyc 4.0. There is no stable submission API or job-polling endpoint here, so keep input, output and service version together. The local helpers need Python 3.10+. GlyTouCan lookup uses requests and network access. Structural glycan shielding calls for a separate GlycoSHIELD install with glycan libraries and GROMACS for SASA analysis; without those the step will not run.

What it helps with

  • scan a protein sequence for canonical N-glycosylation sites
  • knock out one glycosylation site with a point substitution
  • introduce a new N-X-[S/T] site and list every residue changed
  • flag S/T-rich regions before O-GalNAc triage
  • compare antibody glycoforms by site occupancy

Example prompts for the agent

  • Scan this sequence for canonical N-glycosylation sites and list their positions
  • Remove the site at position 1 with a Q substitution and check for overlapping motifs
  • Find S/T-rich regions in this sequence and tell me what to check next

How to install glycoengineering

1Skills CLI (Vercel)

npx skills add K-Dense-AI/scientific-agent-skills --skill glycoengineering -a claude-code

2Manually: put the files in place

~/.claude/skills/glycoengineering/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/glycoengineering/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about glycoengineering

How do I install glycoengineering in Claude Code?

Run npx skills add K-Dense-AI/scientific-agent-skills --skill glycoengineering -a claude-code. The skill lands in ~/.claude/skills/glycoengineering/SKILL.md.

What does glycoengineering do?

It scans for N-glycosylation sites and S/T-rich regions, plans point mutations with a motif rescan, and prepares input for NetNGlyc, NetOGlyc and GlycoSHIELD.

What do the local functions need?

Python 3.10+ and the scripts/ directory on PYTHONPATH. GlyTouCan lookup also needs requests and network access.

Similar skills

All skills for Science and research →

From the same set159

scikit-survival A skill that walks an agent through scikit-survival 0.28.0: building, tuning and auditing survival models on censored data, with metrics that account for censoring. CX Codex +1 2,000installs hugging-science A skill to find and evaluate scientific datasets, models, and interactive demos via the Hugging Science catalog. Use it when you need an ML resource for biology, chemistry, genomics, materials science, climate, or physics. CX Codex +1 1,800installs ginkgo-cloud-lab Assists with protocol selection, input files, pricing checks, and orders for Ginkgo Bioworks Cloud Lab. Useful for RNA synthesis and automated protein expression. CX Codex +1 1,900installs primekg Wraps the PrimeKG precision-medicine CSV in pandas helpers, so the agent can look up typed nodes, neighbours, disease summaries and short paths without leaving the repo. CX Codex +1 1,900installs exa-search Skill for searching scientific and technical content via Exa with text extraction from pages and PDFs in batches. Supports filtering by publication category and academic domains. CX Codex +1 1,700installs autoskill Analyzes Screenpipe activity windows to detect repeating research workflows and drafts new skills for review. Runs strictly on explicit user demand. CX Codex +1 1,700installs nextflow Guides your AI agent in writing, running, and debugging Nextflow DSL2 pipelines and nf-core workflows on HPC or cloud runners. CX Codex +1 1,400installs pathway-enrichment Runs pathway and gene-set enrichment analysis on gene lists or ranked data, finds over-represented GO terms, KEGG, Reactome, MSigDB. Needed after differential expression, CRISPR screens, or clustering. CX Codex +1 1,400installs

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor