Haddock logo

Haddock

Community
makabaka007x
haddock

Create, run, and analyze HADDOCK 2.5 biomolecular docking projects. Use for HADDOCK/HADDOCK2.5, CNS/HADDOCK environment checks, protein-protein, protein-nucleic-acid, protein-ligand, peptide, AIR restraints, active/passive residues, restraint-file management, project setup, cluster scoring, or top-model analysis.

Overview

Publishermakabaka007x
Repositorycadd-skill
Skill namehaddock
Stars
54
Forks
6
Bundled files
3
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 3 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by makabaka007x on GitHub. Read the source before you install it.

Installation

Install the Haddock AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/makabaka007x/cadd-skill/tree/main/skills/haddock
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/makabaka007x/cadd-skill.git /tmp/cadd-skill
mkdir -p .claude/skills
cp -r /tmp/cadd-skill/skills/haddock .claude/skills/haddock
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Haddock in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Haddock on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Haddock is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

HADDOCK 2.5 Skill

概述

HADDOCK (High Ambiguity Driven biomolecular DOCKing) 是一个信息驱动的生物分子对接软件,用于预测蛋白质 - 蛋白质、蛋白质 - 核酸、蛋白质 - 小分子等复合物的三维结构。

环境模板

bash
# HADDOCK 安装路径
HADDOCK_DIR=/path/to/haddock2.5

# 激活环境
conda activate haddock2.5
cd $HADDOCK_DIR
source haddock_configure.sh

# 运行对接
haddock2.5

运行配置

  • CPU 核数、batchmode 和 it0/it1/water 批量大小必须按目标机器资源调整。
  • 不要把某个本地环境的参数当作所有用户默认值。

快速开始

基本命令

/haddock run <example_name>     # 运行示例
/haddock status                 # 查看运行状态
/haddock analyze <run_dir>      # 分析结果
/haddock clean <run_dir>        # 清理结果
/haddock env                    # 检查环境
/haddock examples               # 列出所有示例
/haddock project new <name>     # 创建新项目
/haddock param set <key> <val>  # 修改参数
/haddock restrain create <type> # 创建约束文件

运行第一个对接

bash
/haddock run protein-protein

这将:

  1. 检查并激活 haddock2.5 环境
  2. 进入 protein-protein 示例目录
  3. 应用补丁并启动对接
  4. 后台运行并显示日志

命令参考

1. /haddock env - 环境管理

功能: 检查、激活或显示 HADDOCK 环境状态

bash
/haddock env                    # 完整检查并激活
/haddock env check              # 仅检查状态
/haddock env info               # 显示环境信息

输出示例:

[✓] Conda 环境:haddock2.5 已激活
[✓] HADDOCK 目录:/path/to/haddock2.5
[✓] CNS 可执行文件:存在
[✓] 配置文件:haddock_configure.sh 已加载

2. /haddock examples - 示例管理

功能: 列出、描述或运行内置示例

bash
/haddock examples                    # 列出所有示例
/haddock examples detail <name>      # 显示示例详情
/haddock examples run <name>         # 运行指定示例
/haddock examples clean              # 清理所有示例结果

可用示例列表:

示例名称对接类型系统数据源
protein-protein蛋白 - 蛋白E2A-HPRCSP
protein-dna蛋白-DNA3CRO足迹
protein-ligand蛋白 - 小分子神经氨酸酶结合位点
protein-peptide蛋白 - 多肽1NX1AIR
protein-peptide-ensemble集合平均 PRESUMO-DAXXPRE
protein-protein-em冷冻电镜核糖体EMDB 1884
protein-protein-rdcRDC 约束双泛素RDC
protein-protein-pcsPCS 约束EPS-HotPCS
protein-protein-daniDANI 约束E2A-HPRDANI
protein-trimer同源三聚体1QU9WHISCY
protein-tetramer-CG粗粒化四聚体3GD8C4 对称
solvated-docking溶剂化对接Barnase-BarstarWHISCY
protein-refine-pcsPCS 优化-PCS
refine-complex复合物优化--
protein-ligand-shape形状约束BACE_1模板

3. /haddock run - 运行对接

功能: 启动 HADDOCK 对接运行

bash
/haddock run <example_name>         # 运行内置示例
/haddock run --project <name>       # 运行自定义项目
/haddock run --dir <path>           # 运行指定目录
/haddock run --dry                  # 预演(不实际运行)

运行流程:

  1. 环境检查
  2. 进入目录
  3. 应用补丁(如需要)
  4. 启动 haddock2.5
  5. 显示实时日志

确认提示:

即将运行:protein-protein 示例
预计时间:4-12 小时
模型数量:it0=1000, it1=200, water=200
是否继续?[y/N]

4. /haddock status - 查看状态

功能: 监控运行进度和状态

bash
/haddock status                 # 当前运行状态
/haddock status <run_dir>       # 指定目录状态
/haddock status --watch         # 实时监控
/haddock status --tail          # 显示日志尾部

输出示例:

运行目录:run1
当前阶段:water (3/3)
进度:156/200 模型 (78%)
已用时间:3h 24m
预计剩余:45m
当前能量:Evdw=-42.3, Eelec=-298.1, EAIR=-18.5

5. /haddock analyze - 结果分析

功能: 分析对接结果,生成统计报告

bash
/haddock analyze                        # 分析当前 run1
/haddock analyze <run_dir>              # 分析指定目录
/haddock analyze --full                 # 完整分析(包括 RMSD)
/haddock analyze --cluster              # 仅聚类分析
/haddock analyze --energy               # 仅能量分析
/haddock analyze --report               # 生成报告文件

分析内容:

  • 聚类统计(簇大小、排名)
  • RMSD 分析(i-RMSD, l-RMSD)
  • 能量项统计(VdW, 静电,去溶剂化,AIR)
  • HADDOCK 评分排序

输出示例:

=== 聚类统计 ===
簇 1: 89 模型 (44.5%) - HADDOCK 评分:-92.3 ± 11.2
簇 2: 45 模型 (22.5%) - HADDOCK 评分:-85.1 ± 9.8
簇 3: 33 模型 (16.5%) - HADDOCK 评分:-78.4 ± 12.1

=== 最佳模型 ===
结构 1: cluster_1/model_1.pdb
  HADDOCK 评分:-105.2
  VdW: -48.3 | 静电:-312.5 | AIR: -24.1
  i-RMSD: 1.2Å | l-RMSD: 2.1Å

6. /haddock clean - 清理结果

功能: 删除运行生成的文件和目录

bash
/haddock clean                    # 清理当前 run1
/haddock clean <run_dir>          # 清理指定目录
/haddock clean --all              # 清理所有 run* 目录
/haddock clean --dry              # 预演(显示将删除的内容)

确认提示:

将删除:
- run1/structures/ (约 2.5GB)
- run1/analysis/ (约 150MB)
- run1/haddock.out

保留:
- run1/run.cns
- run1/run.param

是否继续?[y/N]

7. /haddock project - 项目管理

功能: 创建、配置和管理自定义对接项目

bash
/haddock project new <name>           # 创建新项目
/haddock project list                 # 列出所有项目
/haddock project info <name>          # 显示项目信息
/haddock project delete <name>        # 删除项目

创建项目流程:

bash
/haddock project new my_complex
  1. 创建目录结构:

    ~/haddock_projects/my_complex/
    ├── pdb/           # PDB 文件
    ├── restraints/    # 约束文件
    ├── run/           # 运行目录
    └── analysis/      # 分析结果
  2. 生成模板文件:

    • run.param.template
    • air.tbl.template
  3. 准备 PDB 文件(可选):

    • 移除水分子
    • 添加氢原子
    • 修复缺失残基

8. /haddock param - 参数调整

功能: 修改 HADDOCK 运行参数

bash
/haddock param list                 # 列出所有参数
/haddock param show <key>           # 显示参数详情
/haddock param set <key> <value>    # 设置参数
/haddock param reset                # 恢复默认值

常用参数:

参数默认值说明
init_struc1000it0 刚体模型数
it0_struc200it1 半柔性模型数
water_struc200water 优化模型数
clust_struc4聚类大小
iniseed917随机种子
em_kscale15000EM 约束力常数
amb_cool110.0AIR 约束能量常数

示例:

bash
/haddock param set init_struc 500     # 减少采样加速
/haddock param set iniseed 1234       # 改变随机种子
/haddock param set water_struc 0      # 跳过 water 阶段

9. /haddock restrain - 约束文件管理

功能: 创建和编辑约束文件

bash
/haddock restrain create air          # 创建 AIR 约束
/haddock restrain create rdc          # 创建 RDC 约束
/haddock restrain create pcs          # 创建 PCS 约束
/haddock restrain create dani         # 创建 DANI 约束
/haddock restrain edit <file>         # 编辑约束文件
/haddock restrain validate <file>     # 验证约束格式

AIR 约束模板:

! 活性残基 (结合界面)
assign (resid 45 and segid A) (resid 100 and segid B) 2.0 2.0 0.0

! 被动残基 (表面活性)
assign (resid 46 and segid A) (resid 101 and segid B) 2.0 2.0 0.0

三阶段对接流程

it0 - 刚体对接

目的:全局搜索所有可能的结合模式
模型数:1000
时间:~2-4 小时
输出:run1/structures/it0/

it1 - 半柔性优化

目的:允许界面侧链柔性,诱导契合
模型数:200(从 it0 选择)
时间:~4-8 小时
输出:run1/structures/it1/

water - 水优化

目的:显式溶剂中精确优化,水桥稳定
模型数:200(从 it1 选择)
时间:~8-16 小时
输出:run1/structures/water/

运行自定义对接项目

步骤 1: 创建项目

bash
/haddock project new my docking

步骤 2: 准备 PDB 文件

bash
# 将 PDB 文件复制到项目目录
cp /path/to/protein_A.pdb ~/haddock_projects/my_docking/pdb/
cp /path/to/protein_B.pdb ~/haddock_projects/my_docking/pdb/

步骤 3: 生成参数文件

bash
/haddock param generate --project my_docking

步骤 4: 定义约束(可选)

bash
/haddock restrain create air --project my_docking

步骤 5: 运行对接

bash
/haddock run --project my_docking

故障排除

常见问题

1. haddock2.5: command not found

解决:

bash
conda activate haddock2.5
source haddock_configure.sh
2. CNS 执行错误

检查:

bash
ls -la $HADDOCK_DIR/cns1.3/cns.*

修复:

bash
chmod +x $HADDOCK_DIR/cns1.3/cns.*
3. 内存不足

解决: 减少采样数

bash
/haddock param set init_struc 500
/haddock param set it0_struc 100
4. 运行太慢

解决:

  • 检查并行配置
  • 减少 water 阶段模型数
  • 考虑跳过 water 阶段
5. Patch 应用失败

手动应用:

bash
cd run1
patch -p0 -i ../run.cns.patch

结果解读

HADDOCK 评分

HADDOCK 评分 = w_vdw*E_vdw + w_elec*E_elec + w_desolv*E_desolv + w_AIR*E_AIR

常见 water 阶段权重:
- w_vdw = 1.0
- w_elec = 0.2
- w_desolv = 1.0
- w_AIR = 0.1 或 1.0(以 run.cns 为准)

实际分析时,必须以当前 run 的 run.cns 权重为准,不要机械套用默认值。

当前版本推荐分析流程(蛋白-核酸)

当前版本建议把 HADDOCK 分析拆成两层:

  1. 官方脚本基线
    • ana_structures.csh
    • ana_clusters.csh
    • ana_clusters.csh -best 4
    • ana_clusters.csh -best 10 作为可选扩展
  2. 扩展分析
    • contact frequency
    • contact map
    • score-vs-FCC
    • score-vs-interface-RMSD
    • score_boxplots
    • energy_boxplots
    • MD shortlist
    • cluster-grouped PDB export

为什么默认保留 full + best4

  • full-cluster average 是官方 cluster 统计的基本口径,必须保留
  • best4 用来检验“大簇是否被尾部较差模型拖分”,默认建议保留
  • best10 是稳健性扩展项,不是每次都必须跑

一个容易混淆的点是:

  • bar/error bar/boxplot 只能改变 full-cluster 数据的展示方式
  • best4 改变的是 cluster 代表分数的统计口径

所以,bar/boxplot 不能替代 best4。

当前版本图层级建议

正文优先

  • cluster/contact map
  • aptamer_contact_frequency
  • score-vs-FCC
  • score-vs-interface-RMSD
  • score_boxplots

补充材料优先

  • cluster_convergence
  • protein_contact_frequency
  • energy_boxplots

图的解释口径

1. score-vs-FCC
  • 用于看主要 cluster family 的分离程度
  • 主文默认展示 top 5 clusters
  • 如需更紧凑的稿件版,可再收缩到 top 3 clusters
  • 其余模型统一归为 Other
  • 这类图是 contact-based similarity diagnostic
  • 它不等同于 canonical HADDOCK score-vs-RMSD
2. score-vs-interface-RMSD
  • 默认相对 major population cluster representative 计算
  • 用于看各模型相对主解的界面几何偏离
  • 主文默认也展示 top 5 clusters
  • 如需更紧凑的稿件版,可再收缩到 top 3 clusters
  • 若没有实验参考复合物,这张图比 score-vs-RMSD to native 更现实
3. score_boxplots
  • 可以进入正文
  • 作用是展示各体系内部主要 cluster 的 HADDOCK score 分布与离散度
  • 不能用来支持不同体系间绝对 HADDOCK score 的直接横比
  • 主文默认与 score-vs-FCC / score-vs-interface-RMSD 保持一致,也展示 top 5 clusters
4. energy_boxplots
  • 用于拆解 Evdw / Eelec / Eair / Edesolv
  • 更适合作为补充图,而不是主结论图

Contact 的默认定义

当前版本 contact 统计默认定义为:

  • 蛋白链 A 与核酸链 B
  • 任意蛋白-核酸重原子距离 <= 4.0 Å
  • 默认基于 water 阶段最低分 top 20 模型计算接触频率与 contact map

这一定义适合做界面总览;若要区分主簇与替代簇,建议额外画 cluster-specific contact map。

跨体系比较的底线

对于不同长度、不同约束规模的体系(例如全长 aptamer vs 截短 aptamer),默认遵循:

  • 同一 run / 同一 stage 内,不同 cluster 的 HADDOCK score 可以比较
  • 不同体系间的绝对 HADDOCK score 默认不可直接横比

跨体系更应优先比较:

  • cluster 收敛性
  • AIR 违约
  • BSA
  • 蛋白接触区域
  • 核酸接触区域
  • binding mode 是否一致

按 cluster 分组导出 PDB

当前版本的标准交付里,建议把 water 最终结构按 cluster 复制导出,便于人工看结构、挑 MD 候选和做 PyMOL。

默认目录组织:

text
analysis_exports/
├── clusters/
│   └── water/
│       ├── cluster_C1/
│       ├── cluster_C2/
│       └── cluster_manifest.csv
└── md_shortlist/
    ├── <system>_C1_modelXX_*.pdb
    └── md_shortlist_manifest.csv

默认规则:

  • 每个 cluster_Cn/ 目录复制该簇全部 water PDB 副本
  • 文件名保留原始模型编号
  • cluster_manifest.csv 至少包含:
    • stage
    • cluster
    • model_id
    • score
    • pdb_filename
    • source_path
    • is_representative
  • 代表模型默认取该 cluster 的最低分模型
  • md_shortlist/ 默认再复制每个体系的 2 个 MD 候选代表结构

Worked Example:RPN10 / #54 vs #54(T)

当前版本的 worked example 固定如下:

  • #54 = full-length aptamer, 76 nt
  • #54(T) = truncated aptamer, 36 nt
  • #54(T) 1-36 -> #54 21-56
  • 蛋白主界面窗口:116-135

推荐默认结论框架:

  • #54(T):C1 为 major population,C8 为 alternative family
  • #54:C1 为主簇,C4 用于检验末端参与
  • 后续 MD 默认每个体系保留 2 个候选:
    • 1 个主人口簇候选
    • 1 个机制上不同的替代候选

质量标准

等级i-RMSDl-RMSD说明
高精度<1Å<2Å接近晶体结构
中等1-2Å2-5Å可接受的预测
低精度>2Å>5Å需要优化

高级功能

对称性约束

cns
! C2 对称
c2sym_distance = 0.0
c2sym_angle = 180.0

多体对接

bash
# 在 run.param 中设置
N_COMP=3  # 三体
N_COMP=4  # 四体(最多支持 20)

粗粒化模式

bash
# 使用 Martini CG 表示
python $HADDOCK_DIR/cg_tools/aa2cg.py input.pdb > output_CG.pdb

参考文献

  1. Dominguez C, et al. JACS 125, 1731-1737 (2003)
  2. de Vries SJ, et al. Proteins 69, 726-733 (2007)
  3. van Dijk ADJ, et al. NAR 34, 3317-3325 (2006)

资源链接


HADDOCK 2.5 Skill - 最后更新:2026-04-10

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Haddock AI skill do?

Create, run, and analyze HADDOCK 2.5 biomolecular docking projects. Use for HADDOCK/HADDOCK2.5, CNS/HADDOCK environment checks, protein-protein, protein-nucleic-acid, protein-ligand, peptide, AIR restraints, active/passive residues, restraint-file management, project setup, cluster scoring, or top-model analysis.

Why use Haddock on TypingMind?

Because you install it once and use it with any model. Haddock is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Haddock in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/makabaka007x/cadd-skill/tree/main/skills/haddock. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Haddock?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Haddock?

As many as you like. As long as a model supports skills, you can use Haddock with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Haddock AI skill free?

It is published on GitHub by makabaka007x. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇