GSE78961 Processing Pipeline

Publication

Genome medicine (2016) — PMID 27655340

Dataset

Modeling the Neuropathology of Tuberous Sclerosis with Human Stem Cells Reveals a Role for Inflammation and Angiogenic Growth Factors

1. 1

   Linker tags were removed from RNA sequencing and ribosome profiling reads by the FASTX Toolkit, v0.0.13 (http://hannonlab.cshl.edu/fastx_toolkit/)

   RNA-seq v0.0.13 GitHub

   $ Bash example

   # Install FASTX Toolkit (if not already installed)
   # conda install -c bioconda fastx_toolkit
   
   # Placeholder for linker sequence. This needs to be provided based on experimental design.
   # Example linker sequence, replace with actual if known.
   LINKER_SEQUENCE="AGATCGGAAGAG"
   
   # Placeholder for input and output file names.
   INPUT_FILE="raw_reads.fastq"
   OUTPUT_FILE="linker_trimmed_reads.fastq"
   
   # Remove linker tags using fastx_clipper
   # -a: linker sequence to clip
   # -i: input FASTQ file
   # -o: output FASTQ file
   fastx_clipper -a "${LINKER_SEQUENCE}" -i "${INPUT_FILE}" -o "${OUTPUT_FILE}"

   Copy View on GitHub
2. 2

   All reads that mapped to rRNAs, tRNAs or mitochondrial rRNAs were removed, and the remaining reads were mapped to RefSeq (v38) by TopHat v2.0.13.

   TopHat v2.0.13 GitHub

   $ Bash example

   # Install TopHat (if not already installed)
   # conda install -c bioconda tophat=2.0.13
   
   # Define variables
   # Replace with actual paths to your reference files and input reads
   GENOME_FASTA="/path/to/GRCh38.fa" # Example: Human genome GRCh38 (RefSeq v38 implies GRCh38)
   GTF_FILE="/path/to/GRCh38_RefSeq.gtf" # Example: RefSeq annotations for GRCh38
   BOWTIE2_INDEX_PREFIX="/path/to/GRCh38_bowtie2_index/GRCh38" # Prefix for Bowtie2 index files
   INPUT_READS="filtered_reads.fastq" # Reads after removal of r/t/mtRNA mappings
   OUTPUT_DIR="tophat_output"
   NUM_THREADS=8 # Example: Number of threads to use
   
   # Build Bowtie2 index (if not already built)
   # This step is usually done once for a given genome
   # bowtie2-build $GENOME_FASTA $BOWTIE2_INDEX_PREFIX
   
   # Create output directory
   mkdir -p $OUTPUT_DIR
   
   # Map reads to RefSeq (v38) using TopHat
   # -p: number of threads
   # -G: GTF file for known transcripts (improves splice junction detection)
   # $BOWTIE2_INDEX_PREFIX: Path to the Bowtie2 index
   # $INPUT_READS: Input FASTQ file (assumed to be single-end; for paired-end, use -1 and -2)
   tophat2 -p $NUM_THREADS -G $GTF_FILE -o $OUTPUT_DIR $BOWTIE2_INDEX_PREFIX $INPUT_READS

   Copy View on GitHub
3. 3

   Finally all read counts that mapped uniquely to genes were extracted for expression analysis with the help of samtools, v1.1.

   samtools v1.1 GitHub

   $ Bash example

   # Install samtools v1.1
   # conda install -c bioconda samtools=1.1
   
   # Placeholder: input alignment file (BAM)
   INPUT_BAM="aligned_reads.bam"
   
   # Output file for uniquely mapped reads (BAM format)
   OUTPUT_BAM="uniquely_mapped_reads.bam"
   
   # Extract reads that mapped uniquely to genes
   # This command filters the input BAM file to retain only uniquely mapped reads.
   # -F 0x100: Exclude secondary alignments (often indicative of multi-mappers or less reliable alignments)
   # -q 20: Minimum mapping quality of 20 (a common threshold for uniquely mapped reads)
   # -b: Output in BAM format
   # The resulting BAM file (uniquely_mapped_reads.bam) would then be used by downstream
   # expression analysis tools (e.g., featureCounts, htseq-count) to generate gene-level counts.
   samtools view -F 0x100 -q 20 -b "${INPUT_BAM}" > "${OUTPUT_BAM}"
   
   # Index the filtered BAM file (optional, but good practice for downstream tools)
   samtools index "${OUTPUT_BAM}"

   Copy View on GitHub

Tools Used