Part III: Test Data Collection (optional)

We orginally curated this test data, contaiining 6 samples of different conditions (see table below), for the purpose of testing and troubleshooting this pipeline. We also use them for showcasing the input formats accepted by this pipeline, and benchmarking pipeline performance (e.g., Run Time, CPU Time and Max Memory Usage), though these six samples were prepared by downsampling from real data.

	Format	Library Type	Phred Encoding	Species	Number of Reads	Preparation
sample1	FASTQ	Paired-end	Phred+33	human	19,410,373 * 2	Downsampled from real data
sample2	FASTQ	Single-end	Phred+33	mouse	16,005,450	Downsampled from real data
sample3	FASTQ, multiple lanes	Paired-end	Phred+33	human	19,410,373 * 2	Split by lane from sample1
sample4	FASTQ, multiple lanes	Single-end	Phred+33	mouse	16,005,450	Split by lane from sample2
sample5	BAM/SAM (to genome)	Paired-end	Phred+33	human	13,856,075 * 2	Downsampled from real data
sample6	BAM/SAM (to transcriptome)	Single-end	Phred+33	mouse	13,533,162	Downsampled from real data

This is only one situation that you need collect the test data: you would like to walk through this pipeline but do not have your own bulk RAN-seq available. In this case, you can download the test data for one or more samples to get started.

• For those who can access to the pre-built conda environment, the test data is available at: /research_jude/rgs01_jude/groups/yu3grp/projects/software_JY/yu3grp/conda_env/bulkRNAseq_2025/pipeline/testdata.
• For other users, please download them from Zenodo.