RNA-seq_bulk

Run seperately

A. Preprocessing

1. Generating Genome Index (might already exist, check folder with name '_index')

Package used: STAR (Required) Download genome(.fa) and annotation(.gtf) from ncbi genome dataset

Indexing allows the aligner to quickly locate potential matches.

Input: genome annotatioin(.gtf), Reference genome(.fa)

Output: index files

• --runThreadN <NumberOfThreads>
  Number of CPU threads to use (set to available cores).

• --runMode genomeGenerate
  Tells STAR to build genome indices.

• --genomeDir /path/to/genomeDir
  Directory (must exist) where indices will be stored. Needs ≥100 GB free for mammalian genomes.

• --genomeFastaFiles /path/to/genome1.fa [/path/to/genome2.fa ...]
  One or more reference FASTA files (chromosomes can be split across files).

• --sjdbGTFfile /path/to/annotations.gtf
  (Optional) GTF containing exon/junction annotations. Improves splice‐aware mapping.

• --sjdbOverhang <ReadLength − 1>
  Length of genomic sequence around each splice junction. Ideally set to ReadLength−1 (e.g. 99 for 100 bp reads).

Notes:

• Create (mkdir) and clear genomeDir before running.
• STAR stores indices in binary format—avoid manual editing.

Example script

conda activate fastp
STAR --runThreadN 6 \
--runMode genomeGenerate \
--genomeDir INDEX_DIR \
--genomeFastaFiles GENOME_DIR/genome.fa \
--sjdbGTFfile GENOME_DIR/genome.gtf \
--sjdbOverhang 99

2. rRNA removal

Package used: ribodetector

Input: RNA sample transctiptome(fastq.gz)

Output: no rRNA transcriptome(.fa)

conda activate ribodetector
./RMrrna.sh /path/to/fastq_dir /path/to/output_dir

3. Adapter trimming

Package used: fastp

Input: no rRNA transcriptome(.fa)

Output: trimmed RNA transcriptome(.fastq.gz)

conda activate fastp
./april_trim.loop.sh

Notes: Follow prompted instruction to type in the absolute directory

B. Sequence alignment

Package used: STAR

Input: trimmed RNA transcriptome(.fastq.gz)

Output: aligned reads(.bam)

Notes:
INDEX_DIR="~/data/m6A_william/mouseGenome/STAR_mouse_genome_index_ensembl" INDEX_DIR="~/data/m6A_william/mouseGenome/STAR_mouse_genome_index_ncbi"

conda activate fastp
./april_star /path/to/fastq_dir /path/to/output_dir /path/to/index_dir

C. Gene counts

Package used: HTseq-count Input: aligned reads(.bam), genome annotatioin(.gtf)

Output: count file(.txt)

conda activate fastp
./april_count /path/to/fastq_dir /path/to/output_dir /path/to/annotation_dir

Run altogether

./final_version

The script will then prompt you for genome species, input directory, and output directory:

Enter species (mouse/zebrafish) [mouse]: 
Input directory [$pwd] for trasncriptome raw data: 
Output root directory /path/to/your/output/with/run/date: 

Notes:

• Create a parent output directory with the command in the following format mkdir ~/lab_data/samplename_ddmmyy (This should be what you type in for Output root directory)
• This script will create child directory for each step automatically as "step1_RMrrna", "step2_adapter_trim", "step3_alignment", and "step4_gene_count" To learn about the difference in absolute path and relative path, refrence this link: Directory(computing)
• All directory typed in should be in absolute path format. To learn about the difference in absolute path and relative path, refrence this link: Path(computing)