# Scientific Glycomics > 糖鎖構造解析スキル。GlyConnect / GlyGen / GlyCosmos 糖鎖データベース統合検索・糖鎖構造描画・糖タンパク質 グリコシル化部位予測・レクチンバインディング・ 糖鎖マスフラグメンテーション解析パイプライン。 TU 外スキル (直接 REST API + Python ライブラリ)。 - Skill: `nahisaho/scientific-glycomics-2` (Agent Skill) - Install (CLI): `npx skillmds@latest add nahisaho/scientific-glycomics-2` - Raw SKILL.md: https://api.skillmd.com/api/skills/nahisaho/scientific-glycomics-2/raw - Safety review: pending (external: skill-scanner PASS, skillspector PASS) - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Integrations & APIs - Author: nahisaho (https://skillmd.com/u/nahisaho) - Updated: 2026-09-21 - Page: https://skillmd.com/skills/nahisaho/scientific-glycomics-2 --- # Scientific Glycomics GlyConnect / GlyGen / GlyCosmos 糖鎖データベースを統合した 糖鎖構造解析・糖タンパク質グリコサイト予測・レクチン特異性・ 糖鎖 MS フラグメンテーション解析パイプラインを提供する。 ## When to Use - 糖鎖構造を GlyTouCan ID から検索・描画するとき - タンパク質のグリコシル化部位を予測・マッピングするとき - GlyGen/GlyConnect で糖鎖-タンパク質関連を検索するとき - 糖鎖マススペクトルのフラグメンテーション解析を行うとき - レクチン-糖鎖結合特異性を調査するとき --- ## Quick Start ## 1. GlyGen 糖鎖検索 ```python import requests import pandas as pd GLYGEN_API = "https://api.glygen.org" def glygen_glycan_search(glycan_type=None, mass_range=None): """ GlyGen — 糖鎖検索。 Parameters: glycan_type: str | None — 糖鎖タイプ ("N-linked", "O-linked", "GAG" 等) mass_range: tuple | None — (min_mass, max_mass) """ query = {} if glycan_type: query["glycan_type"] = glycan_type if mass_range: query["mass"] = { "min": mass_range[0], "max": mass_range[1]} url = f"{GLYGEN_API}/glycan/search" resp = requests.post(url, json=query, timeout=30) resp.raise_for_status() data = resp.json() results = data.get("results", []) rows = [] for r in results: rows.append({ "glytoucan_ac": r.get("glytoucan_ac", ""), "mass": r.get("mass", 0), "glycan_type": r.get("glycan_type", ""), "composition": r.get( "composition", ""), }) df = pd.DataFrame(rows) print(f"GlyGen search: {len(df)} glycans found") return df def glygen_glycan_detail(glytoucan_ac): """ GlyGen — 糖鎖詳細情報取得。 Parameters: glytoucan_ac: str — GlyTouCan アクセッション """ url = f"{GLYGEN_API}/glycan/detail/{glytoucan_ac}" resp = requests.get(url, timeout=30) resp.raise_for_status() data = resp.json() info = { "glytoucan_ac": data.get("glytoucan_ac", ""), "mass": data.get("mass", 0), "glycan_type": data.get("glycan_type", ""), "iupac": data.get("iupac", ""), "glycoct": data.get("glycoct", ""), "species": [s.get("name", "") for s in data.get("species", [])], "proteins": len(data.get("glycoprotein", [])), } print(f"GlyGen: {glytoucan_ac} → " f"type={info['glycan_type']}, " f"mass={info['mass']:.1f}, " f"proteins={info['proteins']}") return info ``` ## 2. 糖タンパク質グリコサイト検索 ```python def glygen_protein_glycosylation(uniprot_ac): """ GlyGen — タンパク質グリコシル化部位取得。 Parameters: uniprot_ac: str — UniProt アクセッション """ url = f"{GLYGEN_API}/protein/detail/{uniprot_ac}" resp = requests.get(url, timeout=30) resp.raise_for_status() data = resp.json() sites = data.get("glycosylation", []) rows = [] for site in sites: rows.append({ "position": site.get("position", 0), "type": site.get("type", ""), "glytoucan_ac": site.get( "glytoucan_ac", ""), "residue": site.get("residue", ""), "evidence": site.get("evidence", ""), }) df = pd.DataFrame(rows) print(f"GlyGen glycosites: {uniprot_ac} → " f"{len(df)} sites") return df ``` ## 3. 糖鎖 MS フラグメンテーション ```python import numpy as np def glycan_fragmentation(composition, ion_type="[M+Na]+"): """ 糖鎖 MS フラグメンテーション予測。 Parameters: composition: dict — 糖鎖組成 例: {"Hex": 5, "HexNAc": 4, "Fuc": 1, "NeuAc": 2} ion_type: str — イオン種 """ monosaccharide_mass = { "Hex": 162.0528, "HexNAc": 203.0794, "Fuc": 146.0579, "NeuAc": 291.0954, "NeuGc": 307.0903, "Pent": 132.0423, } adducts = { "[M+Na]+": 22.9892, "[M+H]+": 1.0073, "[M+K]+": 38.9632, "[M-H]-": -1.0073, } total_mass = 18.0106 # water for sugar, count in composition.items(): if sugar in monosaccharide_mass: total_mass += (monosaccharide_mass[sugar] * count) adduct = adducts.get(ion_type, 22.9892) precursor_mz = total_mass + adduct # Y-type fragments (reducing end) fragments = [] for sugar, count in composition.items(): if sugar not in monosaccharide_mass: continue for i in range(1, count + 1): loss = monosaccharide_mass[sugar] * i frag_mz = precursor_mz - loss fragments.append({ "type": f"Y (loss {i}x{sugar})", "mz": round(frag_mz, 4), "loss": round(loss, 4), }) df = pd.DataFrame(fragments).sort_values( "mz", ascending=False) print(f"Glycan fragmentation: " f"precursor={precursor_mz:.4f}, " f"{len(df)} fragments") return df ``` ## 4. 糖鎖解析統合パイプライン ```python def glycomics_pipeline(uniprot_ids, output_dir="results"): """ 糖鎖解析統合パイプライン。 Parameters: uniprot_ids: list[str] — UniProt ID リスト output_dir: str — 出力ディレクトリ """ from pathlib import Path output_dir = Path(output_dir) output_dir.mkdir(parents=True, exist_ok=True) # 1) グリコサイトマッピング all_sites = [] for uid in uniprot_ids: sites = glygen_protein_glycosylation(uid) sites["protein"] = uid all_sites.append(sites) if all_sites: sites_df = pd.concat(all_sites, ignore_index=True) sites_df.to_csv( output_dir / "glycosites.csv", index=False) # 2) 糖鎖詳細取得 unique_glycans = set() for df in all_sites: if not df.empty: unique_glycans.update( df["glytoucan_ac"].dropna().unique()) glycan_details = [] for gac in list(unique_glycans)[:50]: if gac: detail = glygen_glycan_detail(gac) if detail: glycan_details.append(detail) if glycan_details: gdf = pd.DataFrame(glycan_details) gdf.to_csv( output_dir / "glycan_details.csv", index=False) print(f"Glycomics pipeline → {output_dir}") return {"sites": sites_df if all_sites else pd.DataFrame()} ``` --- ## パイプライン統合 ``` proteomics-mass-spectrometry → glycomics → pathway-enrichment (LC-MS/MS PTM 同定) (糖鎖構造) (糖鎖パスウェイ) │ │ ↓ protein-structure-analysis ────┘ immunoinformatics (糖鎖結合サイト構造) (抗体グリコシル化) ``` ## パイプライン出力 | ファイル | 説明 | 次スキル | |---------|------|---------| | `results/glycosites.csv` | グリコシル化部位 | → protein-structure-analysis | | `results/glycan_details.csv` | 糖鎖詳細 | → pathway-enrichment | ## ToolUniverse 連携 | TU Key | ツール名 | 連携内容 | |--------|---------|--------| | `glygen` | GlyGen | 糖鎖構造・機能データベース検索 |