Token导航 LogoToken导航TokenDH.com
研究检索只读github未标认证来源可访问许可证需确认审计通过

omicverse-visualization-for-bulk-color-systems-and-single-cell-d用于散装颜色系统和单细胞 d 的 omicverse 可视化

Agent Skill

用于辅助数据整理、表格处理、CSV/Excel 分析、指标计算和图表准备。它适合让 Agent 清洗字段、汇总数据、发现异常、生成统计口径或把分析结果转成可读说明。使用时需要确认数据来源、字段含义和时间范围,避免把样本数据当全量事实;涉及敏感数据、导出文件或批量写回时,应先确认权限和脱敏边界。

总安装

749

周安装

30

GitHub Stars

964

下载量

242
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:omicverse-visualization-for-bulk-color-systems-and-single-cell-d(用于散装颜色系统和单细胞 d 的 omicverse 可视化)
来源仓库:https://github.com/starlitnightly/omicverse
仓库路径:skills/omicverse-visualization-for-bulk-color-systems-and-single-cell-d
安装命令:
npx skills add https://github.com/starlitnightly/omicverse --skill omicverse-visualization-for-bulk-color-systems-and-single-cell-d
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/starlitnightly/omicverse --skill omicverse-visualization-for-bulk-color-systems-and-single-cell-d

简介

omicverse-visualization-for-bulk-color-systems-and-single-cell-d 用于辅助数据整理、表格处理和指标计算,适合清洗字段、汇总数据和生成统计说明。

  • 适用于 CSV/Excel 分析、图表准备及发现数据异常的场景。
  • 使用时需确认数据来源、字段含义和时间范围,避免将样本当作全量事实。
  • 涉及敏感数据或导出文件时,应先确认权限和脱敏边界。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

OmicVerse visualization for bulk, color systems, and single-cell data

Overview

Leverage this skill when a user wants help recreating or adapting plots from the OmicVerse plotting tutorials:

It covers how to configure OmicVerse's plotting style, choose colors from the Forbidden City palette, and generate bulk as well as single-cell specific figures.

Instructions

  1. Set up the plotting environment

- Import omicverse as ov, matplotlib.pyplot as plt, and other libraries required by the user's request (pandas, seaborn, scanpy, etc.). - Call ov.ov_plot_set() (or ov.plot_set() depending on the installed version) to apply OmicVerse's default styling before generating figures. - Load example data via ov.read(...)/ov.pp.preprocess(...) or instruct users to supply their own AnnData/CSV files.

  1. Bulk RNA-seq visuals (t_visualize_bulk)

- Use ov.pl.venn(sets=..., palette=...) to display overlaps among DEG lists (no more than 4 groups). Encourage setting sets as a dictionary of set names → gene lists. - For volcano plots, load the DEG table (result = ov.read('...csv')) and call ov.pl.volcano(result, pval_name='qvalue', fc_name='log2FoldChange',...). Explain optional keyword arguments such as sig_pvalue, sig_fc, palette, and label formatting. - To compare group distributions with box plots, gather long-form data (e.g., from seaborn.load_dataset('tips')) and invoke ov.pl.boxplot(data, x_value=..., y_value=..., hue=..., ax=ax, palette=...). Mention how to adjust figure size, legend placement, and significance annotations.

  1. Color management (t_visualize_colorsystem)

- Introduce the color book via fb = ov.pl.ForbiddenCity() and demonstrate fb.get_color(name='凝夜紫') for specific hues. - Show how to pull predefined palettes (ov.pl.green_color, ov.pl.red_color, etc.) and build dicts mapping cell types/groups to color hex codes. - For segmented gradients, combine colors and call ov.pl.get_cmap_seg(colors, name='custom'), then pass the colormap into Matplotlib/Scanpy plotting functions. - Highlight using these palettes in embeddings: ov.pl.embedding(adata, basis='X_umap', color='clusters', palette=color_dict, ax=ax).

  1. Single-cell visualizations (t_visualize_single)

- Remind users to preprocess AnnData if needed (adata = ov.pp.preprocess(adata, mode='shiftlog|pearson', n_HVGs=2000)). - IMPORTANT - Data validation: Before plotting, always verify that required data exists: # Before plotting by clustering or other categorical variable color_col = 'leiden' # or 'clusters', 'celltype', etc. if color_col not in adata.obs.columns: raise ValueError(f"Column '{color_col}' not found in adata.obs. Available columns: {list(adata.obs.columns)}") # Before plotting embeddings basis = 'X_umap' # or 'X_pca', 'X_tsne', etc. if basis not in adata.obsm.keys(): raise ValueError(f"Embedding '{basis}' not found in adata.obsm. Available embeddings: {list(adata.obsm.keys())}") - For palette optimization, use ov.pl.optim_palette(adata, basis='X_umap', colors='clusters') to auto-generate color schemes when categories clash. - Reproduce stacked proportions with ov.pl.cellproportion(adata, groupby='clusters', celltype_clusters='celltype', ax=ax) and transform into stacked area charts by setting kind='area'. - Showcase compound embedding utilities: - ov.pl.embedding_celltype to place counts/proportions alongside UMAPs. - ov.pl.ConvexHull or ov.pl.contour for highlighting regions of interest. - ov.pl.embedding_adjust to reposition legends automatically. - ov.pl.embedding_density for density overlays, controlling smoothness with adjust. - For spatial gene density, describe the workflow: ov.pl.calculate_gene_density(adata, genes=[...], basis='spatial'), then overlay with ov.pl.embedding(..., layer='gene_density', cmap='...'). - For heatmaps, prefer the Marsilea mainline family: - ov.pl.group_heatmap for grouped expression summaries. - ov.pl.feature_heatmap for cell-level ordered heatmaps. - ov.pl.dynamic_heatmap for pseudotime/lineage heatmaps. - ov.pl.cell_cor_heatmap for group similarity heatmaps. - Treat ov.pl.complexheatmap and ov.pl.marker_heatmap as compatibility entry points for older workflows rather than the default extension surface. - Keep default border=False unless a user explicitly asks for framed panels; this matches current OmicVerse heatmap styling more closely. - For trajectory heatmaps, prefer real inferred pseudotime stored on the AnnData object over synthetic ordering whenever notebook or cached lineage results are available. - Cover additional charts like ov.pl.single_group_boxplot, ov.pl.bardotplot, ov.pl.dotplot, and legacy ov.pl.marker_heatmap, emphasizing input formats (long-form DataFrame vs. AnnData with .obs annotations) and optional helpers such as ov.pl.add_palue for manual p-value annotations.

  1. Finishing touches and exports

- Encourage adding titles, axis labels, and fig.tight_layout() to prevent clipping. - Suggest saving figures with fig.savefig('plot.png', dpi=300, bbox_inches='tight') and documenting color mappings for reproducibility. - Troubleshoot common issues: - Missing AnnData keys: Always validate adata.obs columns and adata.obsm embeddings exist before plotting - Palette names not found: Verify color dictionaries match actual category values - Matplotlib font rendering: When using Chinese characters, ensure appropriate fonts are installed - "Could not find X in adata.obs": Check that clustering or annotation has been performed before trying to visualize results. Use defensive checks to compute missing prerequisites on-the-fly.

Examples

  • "Plot a three-set Venn diagram of overlapping DEG lists and reuse Forbidden City colors for consistency."
  • "Load the dentate gyrus AnnData, color clusters with fb.get_color selections, and render an embedding with adjusted legend placement."
  • "Generate single-cell proportion bar/area plots plus gene-density overlays using OmicVerse helper functions."

References

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

35.98%
按下载量换算87

Claude

27.78%
按下载量换算67

Cursor

18.87%
按下载量换算46

Gemini CLI

9.4%
按下载量换算23

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills