DiffExplorer - Differential Analysis Tool

DiffExplorer is a lightweight statistical differential analysis tool for biomedical researchers with zero coding background. It enables data import, statistical testing, and chart generation — with results ready right before your lab meeting. Just unzip and run!

Features

• Automatic test selection: Based on normality and homogeneity of variance, automatically chooses t-test/ANOVA, Welch's test, or non-parametric tests (Mann‑Whitney U / Kruskal‑Wallis).
• Multiple post‑hoc tests: Supports SNK, LSD, Tukey, Dunnett, Dunn, Nemenyi, Steel, Games‑Howell, and Dunnett's T3.
• Bar charts: Bar charts with significance markers (*, **, ***, ns), exportable as PNG (500 DPI) and vector PDF.
• Bilingual interface: Switch between Chinese and English on the fly.
• Portable & self‑contained: No installation of R or Python required – just unzip and run.
• Runs locally & Data privacy: Your data never leaves your computer – no cloud upload, no privacy concerns.
• Transparent statistics: Full decision trail (normality, variance, final test) exported in the result, with open-source R code.

Statistical Transparency

• Decision trail: For every indicator, Shapiro-Wilk normality and Levene's homogeneity tests are performed. Results are saved in the Normality and Levene sheets.
• Final test recorded: The Results sheet explicitly states the chosen test (e.g., Student's t, Welch t, Mann-Whitney U, ANOVA, SNK, Dunn+BH) in the Test column.
• Fully reproducible: The complete R code is open-source, so every P value can be audited and reproduced independently.

Quick Start

1. Download and extract DiffExplorer.zip.
2. Double‑click DiffExplorer.exe to launch.
3. Import your data (supports .xlsx / .csv / .txt).
4. Click "Run Analysis". (Default test pre-selected, changeable.)
5. View result tables and charts; export with one click.

System Requirements

• Windows 10 / 11 (64‑bit)
• No additional software needed

Download

Option 1 (Recommended): Download from GitHub Releases
https://github.com/JkingYu/DiffExplorer/releases

Option 2: Baidu Cloud (for users in China)
Link: https://pan.baidu.com/s/1IE-oQQ_rIkfZse4bd20Fow?pwd=kyk2
Password: kyk2

Tech Stack

• Python 3.12 (UI & plotting)
• R 4.6.1 (statistical computation)
• PyInstaller (packaging)

License

This project is licensed under the MIT License - see the LICENSE file for details.

Feedback

Report bugs or suggest features via Issues.

---

DiffExplorer - 差异探索分析工具

DiffExplorer 是一款专为零代码基础的生物医学研究者设计的轻量级统计差异分析工具。支持数据导入、统计检验和图表生成，助你在组会前快速获得分析结果。解压即用！

功能特点

• 自动分流检验：根据数据正态性和方差齐性，自动选择 t/ANOVA、Welch 或非参数检验（MW/KW）。
• 多事后检验：支持 SNK、LSD、Tukey、Dunnett、Dunn、Nemenyi、Steel、Games‑Howell、Dunnett's T3。
• 差异柱状图：柱状图带显著性星号(*, **, ***, ns)，支持 PNG（500 DPI）和 PDF 矢量图导出。
• 中英文切换：界面支持中英文，适合国内外用户。
• 绿色免安装：解压即用，无需安装 R 或 Python。
• 本地运行·数据安全：数据无需上传云端，隐私安全。
• 统计流程透明：结果完整记录决策痕迹（正态性、方差齐性、最终检验），R 代码开源可审计。

统计流程透明度

• 决策痕迹：对每个指标均执行 Shapiro-Wilk 正态性检验和 Levene 方差齐性检验，结果分别保存在 Normality 和 Levene 工作表中。
• 最终检验记录：Results 工作表的 Test 列明确写出所用检验（如 Student's t、Welch t、Mann-Whitney U、ANOVA、SNK、Dunn+BH）。
• 完全可复现：R 统计代码完全开源，每个 P 值均可独立审计和复现。

快速开始

1. 下载并解压 DiffExplorer.zip。
2. 双击 DiffExplorer.exe 启动。
3. 导入数据（支持 .xlsx / .csv / .txt）。
4. 点击“运行分析”。(默认方法，可调整）
5. 查看结果表格和图表，一键导出。

系统要求

• Windows 10 / 11（64位）
• 无需安装任何额外软件

下载

方式一（推荐）：从 GitHub Releases 下载
https://github.com/JkingYu/DiffExplorer/releases

方式二：百度网盘（国内用户加速）
通过网盘分享的文件：DiffExplorer.zip
链接: https://pan.baidu.com/s/1IE-oQQ_rIkfZse4bd20Fow?pwd=kyk2
提取码: kyk2

技术栈

• Python 3.12（UI 与绘图）
• R 4.6.1（统计计算）
• PyInstaller（打包）

许可证

本项目采用 MIT 许可证，详见 LICENSE 文件。

反馈

欢迎通过 Issues 提交 bug 报告或建议。