Readme.Rmd

---
title: "Readme"
output: github_document
date: "`r Sys.Date()`"
---

# Juselius_backdoor

![](coauthorship_plot.png)

**Juselius_backdoor** is an R pipeline for constructing and analyzing coauthorship networks of **Sigrid Juselius Foundation** grantees using **OpenAlex author IDs**. It extracts authors from a text file, resolves their OpenAlex IDs, retrieves their publications, and builds a weighted coauthorship network.

[View Interactive version of this plot](https://jafarilab.github.io/Juselius_backdoor/interactive_plot.html)
---

## Features

- Extracts grantee names from text files.  
- Resolves authors to **OpenAlex IDs**.  
- Collects publication data for each author.  
- Builds a weighted coauthorship network.  
- Computes network metrics:  
  - Degree (number of collaborators)  
  - Weighted degree  
  - Betweenness (key connectors)  
  - Eigenvector centrality (influence)  
- Detects communities using the **Louvain algorithm**.  
- Generates visualizations:  
  - Network graph with community coloring  
  - **Scatter plot of one-time vs. total coauthorship**  
  - Histograms of centrality measures  
- Exports networks for **Gephi** analysis (`.gml` format).  

---

## Installation

```{r, echo=T, eval=FALSE,warning=FALSE}
# Install required packages
install.packages(c(
  "httr", "jsonlite", "dplyr", "igraph", "ggraph", 
  "tidygraph", "ggplot2", "ggrepel", "viridis"
))
```

---

## 🚀 Usage

1. Place your grantee text file in the project folder (e.g., `jus2026adult.txt`).

2. Run the pipeline R scripts in order:

   * Extract author names
   * Search OpenAlex IDs
   * Retrieve papers
   * Build the coauthorship network
   * Compute network metrics
   * Visualize results


## 📊 Output

* Weighted coauthorship network (`.gml`) for Gephi
* Network metrics per author (`netSummary`)
* Visualizations of network structure and centrality


Top 10 Researchers (by weighted degree):

```{r, echo=F}
library(knitr)
netSummary <- read.csv("netSummary.csv")
top10 <- netSummary[order(-netSummary$wdegree), ][1:10, ]
top10_display <- top10[, c("author", "degree", "wdegree", "betweenness", "community", "primary_neighbor_MaxW")]
kable(top10_display, format = "markdown")
```

Top 3 Researchers per Community:
```{r, echo=F}
library(knitr)

if(file.exists("top3_per_community.csv")){
  top3_per_community <- read.csv("top3_per_community.csv")
  top3each <- top3_per_community[order(top3_per_community$community), ]
  top3each <- top3each[, c("author", "wdegree", "community")]
  kable(top3each, format = "markdown")
} else {
  cat("File top3_per_community.csv not found. Run the pipeline first.")
}
```


[View Interactive Co-authorship Network](https://jafarilab.github.io/Juselius_backdoor/network.html)
---

## ⚠️ Notes & Limitations
- Author matching is based on name search and may introduce minor errors
- OpenAlex data is continuously updated → results may vary between runs
- API rate limits and connectivity can affect data retrieval
- Some manual validation was performed, but results are not error-free

## 📌 Interpretation Notes
- High degree + low wdegree → broad but shallow collaboration
- Low degree + high wdegree → deep collaboration with few partners
- Primary neighbor → strongest collaboration link (potential entry point into a research cluster)

## License

MIT License – free to use and modify.

```

This is a fully self-contained **R Markdown README**.  

If you want, I can also add a **“Quick Start” code chunk** at the top so users can generate the scatter plot in **one go**, which is very handy for GitHub. Do you want me to do that?
```