feat: add literature templates and update deployment

- Add 文献阅读.md template for literature notes
- Add 科研读图指南.md main chart reading guide
- Update deploy_workflow_scripts to include template files
This commit is contained in:
Research Assistant 2026-04-21 22:11:48 +08:00
parent 9cadab68ad
commit 3447f0279d
3 changed files with 153 additions and 0 deletions

View file

@ -0,0 +1,47 @@
---
title: " {{标题}} "
year: {{年份}}
type: {{类型}}
journal: " {{期刊}} "
impact_factor: {{影响因子}}
category: {{自己的分类}}
tags:
- 文献阅读
{{#if 标记}} - {{标记}} {{/if}}
keywords: [ {{关键词}} ]
pdf_link: "[[{{上传PDF}}]]"
---
# {{标题}}
## 📄 文献基本信息
{{文献基本信息}}
---
## 💡 文献内容总结
{{文献内容总结}}
---
## 🚀 未来方向
{{未来方向}}
---
## 📝 个人备注与补充
**备注:** {{#if 备注}}
{{备注}}
{{else}}
(无备注)
{{/if}}
**补充说明:**
{{#if 补充}}
{{补充}}
{{else}}
(无补充内容)
{{/if}}
---
**附件:** [[{{上传PDF}}]]

View file

@ -0,0 +1,81 @@
# 科研读图指南
## 使用方式
在精读 **Figure-by-Figure** 时,识别图表类型,然后加载对应子指南。
**识别方法**
1. 看标题/caption中的关键词
2. 看图像本身的视觉特征(坐标轴、颜色、几何图形)
3. 如果不确定,参考下方的「图表类型索引」
---
## 通用检查清单(适用于所有图表)
在深入类型-specific分析前先回答这6个问题
- [ ] **类型匹配**:图表类型是否匹配作者声称的分析方法?(例如:说做差异分析却用散点图而非箱式图)
- [ ] **样本量**n值是否标注每个点/柱/箱代表多少样本?
- [ ] **统计方法**p值、检验方法、置信区间是否明确
- [ ] **颜色图例**:是否自解释?色盲友好吗?
- [ ] **坐标轴**单位、刻度、截断是否合理警惕Y轴截断造成的视觉放大
- [ ] **视觉误导**:是否存在不连续刻度、面积误导、三维透视扭曲等?
---
## 图表类型索引
| 图表类型 | 识别关键词 | 子指南文件 |
|---------|-----------|-----------|
| 箱式图 / 小提琴图 | boxplot, violin, quartile, median, IQR | [[读图指南/箱式图与小提琴图]] |
| 热图 / 聚类图 | heatmap, clustering, dendrogram, z-score | [[读图指南/热图与聚类图]] |
| PCA / t-SNE / UMAP | principal component, embedding, dimensionality reduction | [[读图指南/降维图(PCA-tSNE-UMAP)]] |
| ROC / PR 曲线 | ROC, AUC, sensitivity, specificity, precision-recall | [[读图指南/ROC与PR曲线]] |
| 生存曲线 | Kaplan-Meier, survival, hazard ratio, log-rank | [[读图指南/生存曲线]] |
| 火山图 / 曼哈顿图 | volcano plot, log2 fold change, -log10 p, GWAS | [[读图指南/火山图与曼哈顿图]] |
| 散点图 / 气泡图 | scatter plot, bubble, correlation, regression | [[读图指南/散点图与气泡图]] |
| 条形图 / 误差棒 | bar chart, SEM, SD, confidence interval | [[读图指南/条形图与误差棒]] |
| GSEA 富集图 | enrichment plot, ES, NES, leading edge | [[读图指南/GSEA富集图]] |
| 森林图 / Meta 分析 | forest plot, meta-analysis, odds ratio, I2 | [[读图指南/森林图与Meta分析]] |
| 雷达图 / 漏斗图 | radar chart, funnel plot, publication bias | [[读图指南/雷达图与漏斗图]] |
| 网络图 / 通路图 | network, pathway, PPI, interaction | [[读图指南/网络图与通路图]] |
| 桑基图 / 弦图 | Sankey, chord, alluvial, flow | [[读图指南/桑基图与弦图]] |
| 蛋白质结构图 | protein structure, PDB, Cryo-EM, X-ray | [[读图指南/蛋白质结构图]] |
---
## 进阶:什么时候需要子指南?
**所有 14 类图表均有子指南**。根据复杂度选择阅读深度:
**复杂图表**(强烈建议读子指南):
- 热图(聚类、标准化、注释层)
- PCA/t-SNE/UMAP方差解释、批次效应
- ROC/PR曲线截断点、置信区间
- 生存曲线(删失、比例风险假设)
- GSEA 富集图ES曲线、Leading Edge、FDR阈值
- 森林图(异质性、权重、发表偏倚)
- 蛋白质结构图(分辨率、界面、动态性)
**中等复杂度**(建议读子指南):
- 箱式图/小提琴图(统计检验选择、异常值处理)
- 火山图(多重检验校正、效应量阈值)
- 散点图/气泡图(过绘、假相关、气泡大小误导)
- 条形图/误差棒SEM vs SD、Y轴截断、多重比较
- 网络图/通路图hairball、预测边、Hub偏倚
**相对简单但仍建议参考**
- 雷达图/漏斗图(维度标准化、发表偏倚判断)
- 桑基图/弦图(流量守恒、路径追踪、方向性)
**通用清单足够**(基础检查):
- 简单的饼图、面积图、折线图
---
## References
- Weissgerber et al. (2015). *Beyond bar and line graphs*. PLOS Biology. [DOI:10.1371/journal.pbio.1002128](https://doi.org/10.1371/journal.pbio.1002128)
- Weissgerber et al. (2019). *Reveal, don't conceal*. Circulation. [DOI:10.1161/CIRCULATIONAHA.118.037777](https://doi.org/10.1161/CIRCULATIONAHA.118.037777)
- Krzywinski & Altman (2013). *Visualizing samples with box plots*. Nature Methods. [DOI:10.1038/nmeth.2813](https://doi.org/10.1038/nmeth.2813)

View file

@ -513,6 +513,31 @@ def deploy_workflow_scripts(vault_path: Path, agent_key: str, agent_config: dict
fail_count += 1
print_success(f"Deployed {len(chart_files)} chart reading guides")
# Deploy template files
template_files = [
("99_System/Template/文献阅读.md",
f"{paths['template_path']}/文献阅读.md"),
("99_System/Template/科研读图指南.md",
f"{paths['template_path']}/科研读图指南.md"),
]
for src_rel, dst_rel in template_files:
src_path = repo_root / src_rel
dst_path = vault_path / dst_rel
if src_path.exists():
try:
dst_path.parent.mkdir(parents=True, exist_ok=True)
shutil.copy2(src_path, dst_path)
print_success(f"Deployed template: {src_rel}")
success_count += 1
except Exception as e:
print_error(f"Failed to deploy template {src_rel}: {e}")
fail_count += 1
else:
print_warning(f"Template not found (skipping): {src_rel}")
fail_count += 1
print(f"\nDeployment summary: {success_count} succeeded, {fail_count} failed")
return fail_count == 0