← Back to Skills Marketplace
133
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install bubble-plot
Description
在需要基于基因表达数据画气泡图（dot plot）时使用。适用于单细胞 RNA-seq、空间转录组等场景，按细胞类型 × 癌种展示基因表达比例和均值。
README (SKILL.md)
Skill: 气泡图生成器 (bubble-plot)
基于基因表达矩阵数据，生成按细胞类型 × 癌种分组的气泡图（dot plot）。
何时使用
- 用户上传 TSV/CSV 表达数据文件，并要求画气泡图、dot plot、泡泡图
- 用户说"画气泡图"、"bubble plot"、"dot plot"并提供了数据文件
- 用户提供数据文件和基因列表，要求可视化表达模式
常见触发词
- 画气泡图
- bubble plot
- dot plot
- 泡泡图
- 气泡图可视化
- 表达气泡图
执行清单
1. 检查数据与环境
- 确认用户提供了数据文件（TSV/CSV）
- 确认文件已下载到工作目录
- 确认 Python 3 + pandas + matplotlib 已安装（缺失时
pip3 install --break-system-packages pandas matplotlib）
2. 了解数据结构
读取文件前几行，识别以下关键列：
| 角色 | 识别关键词 | 示例列名 |
|---|---|---|
| 细胞类型 | majorCluster, cell_type, cluster |
majorCluster |
| 组织类型 | tissue, class, condition, group |
tissue |
| 癌种 | cancerType, cancer_type, disease |
cancerType |
| 基因表达 | 所有数值型非元数据列 | FOLR1, TACSTD2, MET ... |
- 如果用户指定了基因列表，只画指定的
- 如果用户未指定，自动检测所有数值型非元数据列
- 将
Adjacent等正常组织名称统一映射为Normal
3. 运行脚本
# 基本用法 — 自动检测所有基因
python3 skills/bubble-plot/bubble_plot.py \x3Cdata.tsv>
# 只画指定基因
python3 skills/bubble-plot/bubble_plot.py \x3Cdata.tsv> --genes FOLR1 TACSTD2 MET
# 自定义列名（如果自动检测失败）
python3 skills/bubble-plot/bubble_plot.py \x3Cdata.tsv> --genes FOLR1 MET --cell-col CellType --cancer-col Disease
# 不按组织类型拆分
python3 skills/bubble-plot/bubble_plot.py \x3Cdata.tsv> --no-split-tissue
# 自定义输出目录和分辨率
python3 skills/bubble-plot/bubble_plot.py \x3Cdata.tsv> -o my_plots --dpi 600
4. 输出文件
- 默认输出到
bubble_plots/目录 - 每个基因 × 组织类型 = 1 个 PNG（300dpi）+ 1 个 PDF
- 文件命名：
{基因}_dotplot_{Normal|Tumor}.{png|pdf}
5. 向用户报告结果
- 列出生成的文件清单和大小
- 如果有基因在数据中不存在，明确告知并跳过
- 展示 1-2 张关键图的预览
参数说明
| 参数 | 默认值 | 说明 |
|---|---|---|
input |
必填 | 输入 TSV 或 CSV 文件路径 |
--genes / -g |
自动检测 | 要画的基因名列表 |
--cell-col |
自动检测 | 细胞类型列名 |
--cancer-col |
自动检测 | 癌种/疾病列名 |
--tissue-col |
自动检测 | 组织类型列名 |
--split-tissue |
True | 按组织类型拆分画图 |
--no-split-tissue |
- | 不拆分，所有数据画一张 |
--outdir / -o |
bubble_plots |
输出目录 |
--dpi |
300 | PNG 分辨率 |
图例说明
- 气泡颜色：标准化平均表达量（0-1，红=高，蓝=低，RdBu_r 色谱）
- 气泡大小：阳性细胞百分比（该细胞类型中表达量 > 0 的比例）
自检
- 数据文件是否已确认下载到本地
- 基因列名是否与数据中一致（区分大小写）
tissue列中Adjacent是否已映射为Normal- 输出是否包含 PNG 和 PDF 两种格式
- 缺失的基因是否已告知用户
Usage Guidance
This skill appears to do what it says: generate bubble/dot plots from a TSV/CSV gene-expression matrix. Before running it: (1) review the included bubble_plot.py (already present) and confirm you trust it — it only reads the provided file and writes PNG/PDF outputs; (2) avoid running the suggested pip command with --break-system-packages on a production/system Python — prefer a virtual environment (python -m venv) or conda to install pandas/matplotlib; (3) ensure the input file does not contain sensitive data you don't want processed or stored on the machine where you run the script; (4) run the skill in an isolated environment if you have any doubt about package changes or file permissions. If you need the agent to run this automatically, note it can execute the local script but has no network exfiltration code included.
Capability Analysis
Type: OpenClaw Skill
Name: bubble-plot
Version: 1.0.0
The bubble-plot skill is a legitimate bioinformatics tool designed to visualize gene expression data from TSV/CSV files using pandas and matplotlib. The Python script (bubble_plot.py) and instructions (SKILL.md) focus entirely on data processing and plot generation without any indicators of data exfiltration, malicious execution, or prompt injection.
Capability Assessment
Purpose & Capability
Name/description, SKILL.md, and the included Python script are consistent: the script reads a TSV/CSV expression matrix, auto-detects metadata and numeric gene columns, and writes PNG/PDF bubble plots. Requiring python3 (and pandas/matplotlib) is proportional to the stated purpose.
Instruction Scope
SKILL.md restricts actions to checking for a local input file, ensuring Python libraries are installed, running the included script, and returning generated plot files. One operational note: the suggested pip command uses --break-system-packages (modifies system Python installation), which may be undesirable — using a virtualenv/conda environment would be safer. Otherwise the instructions do not attempt to read unrelated files or transmit data externally.
Install Mechanism
There is no install spec (instruction-only skill) and the code is included. The script depends on common Python libraries (pandas, matplotlib) and there are no downloads from external URLs or archive extraction steps in the skill package itself.
Credentials
The skill declares no environment variables, no credentials, and no config paths. The code only reads the user-supplied input file path and writes output images, which is appropriate for its function.
Persistence & Privilege
always is false and the skill does not request persistent or elevated platform privileges. It does not attempt to modify other skills or system-wide agent settings.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install bubble-plot - After installation, invoke the skill by name or use
/bubble-plot - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
plot_test
Metadata
Frequently Asked Questions
What is bubble_plot?
在需要基于基因表达数据画气泡图（dot plot）时使用。适用于单细胞 RNA-seq、空间转录组等场景，按细胞类型 × 癌种展示基因表达比例和均值。 It is an AI Agent Skill for Claude Code / OpenClaw, with 133 downloads so far.
How do I install bubble_plot?
Run "/install bubble-plot" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is bubble_plot free?
Yes, bubble_plot is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does bubble_plot support?
bubble_plot is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created bubble_plot?
It is built and maintained by FmaxY (@fmaxy); the current version is v1.0.0.
More Skills