ViralCellDetector

ViralCellDetector is an R-based pipeline designed to detect viral contamination in RNA-seq samples in any host sample. It accepts raw RNA-seq FASTQ files as input and outputs a list of putative viruses identified in the sample. The final results can be visualized using genome browsers such as IGV (Integrative Genomics Viewer).

---

📦 Installation

git clone https://github.com/Bin-Chen-Lab/ViralCellDetector
cd ViralCellDetector

---

🚀 Usage

Step 1: Prepare the Reference Genomes

Before running the detection script, you need to download the host genome and annotation files by executing the provided shell script.

1. Edit the Genome_file.txt to include the appropriate FTP links for your species' genome and annotation files. For example, for the human genome:

ftp://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_human/release_29/GRCh38.primary_assembly.genome.fa.gz  
ftp://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_human/release_29/gencode.v29.annotation.gtf.gz

2. Then run:

bash Genome_index.sh Genome_file.txt

This will:

• Download the host genome and annotation files
• Download the viral genomes from NCBI
• Build STAR and BWA index files for alignment

---

Step 2: Prepare Input Files

1. Create a folder named fastq/ inside the ViralCellDetector directory and move your FASTQ files into this folder.

2. Create a file named sample_input.txt containing the base names (without _1.fastq or _2.fastq suffixes) of your paired-end FASTQ files.
   Example entry for a sample:

   input_file
   

If the paired-end files are input_file_1.fastq and input_file_2.fastq, only write input_file in the sample_input.txt.

---

Step 3: Run the Pipeline

After the setup, execute the main R script:

Rscript ViralCellDetector.R sample_input.txt

---

🧬 Input

• Raw RNA-seq FASTQ files (preferably rRNA-depleted, though poly-A data can also be used)
• A text file (sample_input.txt) listing sample base names
• The fastq/ directory should contain all input FASTQ files

---

📤 Output

• A summary file for each sample listing:
  • Putative virus names
  • Genome size
  • Number of mapped reads
  • Covered genomic regions

The viral mapping file can be loaded into IGV or other genome browsers for visualization.

---

⚠️ Notes

• Ensure high-quality sequencing reads for optimal detection.
• Poly-A enriched data may result in reduced viral diversity compared to rRNA-depleted data.

---

🧩 Dependencies

Make sure the following tools and R packages are installed before running the script:

🔧 Command-line Tools

• samtools
• STAR
• BWA

📦 R Packages

Install these via CRAN if not already available:

install.packages(c("dplyr", "data.table", "tidyr"))

---

🧪 Example

To run the example test case:

bash Run_test_ViralCellDetector.sh

---

👥 Contributors

ViralCellDetector is developed and maintained by the Bin Chen Lab.

For questions or suggestions, please contact:

• Rama Shankar, PhD – ramashan@msu.edu
• Bin Chen, PhD (PI) – Chenbi12@msu.edu