GSE80664 Processing Pipeline

Publication

Cell reports (2016) — PMID 27239029

Dataset

Transcriptome-wide mRNA alterations in Streptozotocin induced Type I diabetic mouse heart

1. 1

   Base calling was performed with CASAVA 1.8.2

   CASAVA v1.8.2

   $ Bash example

   # Install CASAVA 1.8.2 (typically part of Illumina's instrument software)
   # This is a conceptual command for base calling and demultiplexing with CASAVA 1.8.2.
   # Replace /path/to/illumina/run_directory with the actual path to your Illumina run folder.
   # Replace /path/to/output_directory with the desired output location for FASTQ files.
   # Replace /path/to/SampleSheet.csv with the actual path to your sample sheet.
   
   # Step 1: Configure the base calling and demultiplexing
   # This command generates a Makefile in the output directory.
   configureBclToFastq.pl --input-dir /path/to/illumina/run_directory/Data/Intensities/BaseCalls \
                              --output-dir /path/to/output_directory \
                              --sample-sheet /path/to/SampleSheet.csv \
                              --no-eamss # Optional: disable EAMSS (Error-Aware Multi-Sample Splitting)
   
   # Step 2: Execute the base calling and demultiplexing using the generated Makefile
   # Navigate to the output directory where the Makefile was created
   # cd /path/to/output_directory
   make -j 8 # Use an appropriate number of parallel jobs (e.g., number of CPU cores)

   Copy
2. 2

   Sequences were trimmed then aligned to the mm9 mouse genome using TOPHAT

   TopHat v2.1.1 (Inferred with models/gemini-2.5-flash) GitHub

   $ Bash example

   # Install TopHat (example using Conda)
   # TopHat requires Bowtie2 and Samtools as dependencies.
   # conda install -c bioconda tophat bowtie2 samtools
   
   # --- Prepare Reference Genome and Annotation ---
   # 1. Download the mm9 reference genome FASTA file (e.g., from UCSC).
   #    Example: wget -P /path/to/genome/ http://hgdownload.soe.ucsc.edu/goldenPath/mm9/bigZips/mm9.fa.gz
   #    gunzip /path/to/genome/mm9.fa.gz
   
   # 2. Build the Bowtie2 index for mm9.
   #    bowtie2-build /path/to/genome/mm9.fa /path/to/bowtie_indexes/mm9
   
   # 3. Download a GTF annotation file for mm9 (e.g., from UCSC or Ensembl).
   #    Example: wget -P /path/to/annotations/ http://hgdownload.soe.ucsc.edu/goldenPath/mm9/bigZips/genes/mm9.ncbiRefSeq.gtf.gz
   #    gunzip /path/to/annotations/mm9.ncbiRefSeq.gtf.gz
   
   # --- Define Variables ---
   # Replace with actual paths to your files
   BOWTIE_INDEX="/path/to/bowtie_indexes/mm9" # Path to the Bowtie2 index prefix for mm9
   GTF_FILE="/path/to/annotations/mm9.gtf"   # Path to the mm9 GTF annotation file
   TRIMMED_READS_R1="trimmed_reads_R1.fastq" # Path to the trimmed forward reads (FASTQ)
   TRIMMED_READS_R2="trimmed_reads_R2.fastq" # Path to the trimmed reverse reads (FASTQ) - omit if single-end
   OUTPUT_DIR="tophat_alignment_mm9"         # Directory for TopHat output
   NUM_THREADS=8                              # Number of CPU threads to use
   
   # --- Run TopHat Alignment ---
   # -p: Number of threads
   # -G: GTF annotation file (essential for splice junction discovery in RNA-seq)
   # -o: Output directory
   # The Bowtie2 index prefix and input FASTQ files are positional arguments.
   # For paired-end reads, provide both R1 and R2 files.
   # For single-end reads, provide only the R1 file.
   tophat -p "${NUM_THREADS}" -G "${GTF_FILE}" -o "${OUTPUT_DIR}" "${BOWTIE_INDEX}" "${TRIMMED_READS_R1}" "${TRIMMED_READS_R2}"
   

   Copy View on GitHub
3. 3

   Probabilities of isoform abundances were computed using MISO

   MISO v0.5.3 (Inferred with models/gemini-2.5-flash)

   $ Bash example

   # Install MISO (if not already installed)
   # conda install -c bioconda miso
   
   # Placeholder for MISO events index directory (e.g., pre-built for a specific genome like hg38)
   # This directory contains the indexed alternative splicing events (e.g., from UCSC knownGene, Ensembl)
   # Example: MISO_EVENTS_INDEX_DIR="/path/to/miso_events/hg38_v2"
   MISO_EVENTS_INDEX_DIR="/path/to/miso_events_index"
   
   # Placeholder for input RNA-Seq alignment file (BAM format)
   INPUT_BAM="sample_aligned.bam"
   
   # Placeholder for output directory where MISO results will be stored
   OUTPUT_DIR="miso_output"
   
   # Create the output directory if it does not exist
   mkdir -p "${OUTPUT_DIR}"
   
   # Run MISO to compute probabilities of isoform abundances
   # This command will generate various output files, including posterior distributions
   # and summary statistics for isoform usage, from which probabilities can be derived.
   miso --run "${MISO_EVENTS_INDEX_DIR}" "${INPUT_BAM}" --output-dir "${OUTPUT_DIR}"
   

   Copy
4. 4

   Data was filtered based on a Bayes factor of at least 1

   awk (Inferred with models/gemini-2.5-flash) vN/A

   $ Bash example

   # Assuming 'input.tsv' is a tab-separated file where the Bayes factor is in the 5th column.
   # Adjust the column number ($5) and delimiter (-F'\t') as per your input file format.
   # If your file has a header and you want to preserve it, use:
   # (head -n 1 input.tsv && awk -F'\t' 'NR > 1 && $5 >= 1' input.tsv) > filtered_output.tsv
   awk -F'\t' '$5 >= 1' input.tsv > filtered_output.tsv

   Copy
5. 5

   Additional details are included in Verma, S.K., Deshmukh, V., Liu, P., Nutter, C.A., Espejo, R., Hung, M.L., Wang, G.S., Yeo, G.W., and Kuyumcu-Martinez, M.N. (2013).

   R v3.x

   $ Bash example

   # Install R (if not already available)
   # conda install -c r r-base
   
   # This is a placeholder command as the specific R script and parameters
   # are not detailed in the description, which only references a publication.
   # The publication describes a novel role for RBM3 in translation regulation,
   # which would likely involve various bioinformatics analyses in R.
   # Replace 'my_analysis_script.R', 'input_data.txt', and 'output_results.txt'
   # with actual script and file names relevant to the specific analysis.
   Rscript my_analysis_script.R input_data.txt output_results.txt

   Copy
6. 6

   Reactivation of fetal splicing programs in diabetic hearts is mediated by protein kinase C signaling.

   rMATS (Inferred with models/gemini-2.5-flash) v4.1.2 GitHub

   $ Bash example

   # Install rMATS (example using conda)
   # conda create -n rmats_env python=3.8
   # conda activate rmats_env
   # pip install rmats-turbo
   
   # Define input files and reference
   # Placeholder for human genome (GRCh38) and GENCODE annotation
   GENOME_FASTA="/path/to/human_genome/GRCh38.p14.genome.fa"
   GTF_ANNOTATION="/path/to/human_genome/gencode.v44.annotation.gtf"
   
   # Files listing paths to BAM files for each group
   # In a real scenario, these files would be populated with actual paths to aligned RNA-seq BAMs.
   BAM_LIST_DIABETIC="diabetic_bams.txt"
   BAM_LIST_CONTROL="control_bams.txt"
   OUTPUT_DIR="rmats_output_diabetic_vs_control"
   
   # Create placeholder BAM list files for demonstration
   echo "/path/to/aligned_reads/diabetic_sample1.bam" > $BAM_LIST_DIABETIC
   echo "/path/to/aligned_reads/diabetic_sample2.bam" >> $BAM_LIST_DIABETIC
   echo "/path/to/aligned_reads/control_sample1.bam" > $BAM_LIST_CONTROL
   echo "/path/to/aligned_reads/control_sample2.bam" >> $BAM_LIST_CONTROL
   
   # Run rMATS for differential splicing analysis
   # This command compares splicing events between diabetic and control heart samples.
   # Parameters like readLength and libType should be adjusted based on actual experimental data.
   rmats.py --b1 $BAM_LIST_DIABETIC \
            --b2 $BAM_LIST_CONTROL \
            --gtf $GTF_ANNOTATION \
            --readLength 100 \
            --nthread 8 \
            --tmp rmats_tmp \
            --od $OUTPUT_DIR \
            --task diff \
            --libType fr-firststrand
   

   Copy View on GitHub
7. 7

   The Journal of biological chemistry 288, 35372-35386.

   Unknown (Inferred with models/gemini-2.5-flash) vUnknown

   $ Bash example

   # No specific bioinformatics tool or command could be inferred from the provided publication reference.
   # The description refers to a scientific publication, not a bioinformatics step or tool.

   Copy
8. 8

   processed data files format and content: Proccessed data is MISO analysis of skipped exons in comparisions between control and diabetic samples.

   MISO vNot specified (Inferred with models/gemini-2.5-flash)

   $ Bash example

   # Install MISO (if not already installed)
   # pip install miso
   
   # Define variables
   # MISO annotations for skipped exons (e.g., derived from a GFF3 file for a specific genome assembly like hg38).
   # This file needs to be pre-built using the 'index_gff' script from MISO.
   MISO_ANNOTATIONS="path/to/miso_annotations_se.gff3"
   
   # Input BAM files directories for control and diabetic samples.
   # These directories should contain the aligned RNA-Seq BAM files for each group.
   CONTROL_BAM_DIR="path/to/control_bams"
   DIABETIC_BAM_DIR="path/to/diabetic_bams"
   
   # Output directories for MISO quantification results for each sample group
   OUTPUT_DIR_CONTROL_QUANT="miso_quant_control"
   OUTPUT_DIR_DIABETIC_QUANT="miso_quant_diabetic"
   
   # Output directory for the MISO comparison results
   OUTPUT_DIR_COMPARISON="miso_comparison_output"
   
   # Create output directories if they don't exist
   mkdir -p "${OUTPUT_DIR_CONTROL_QUANT}"
   mkdir -p "${OUTPUT_DIR_DIABETIC_QUANT}"
   mkdir -p "${OUTPUT_DIR_COMPARISON}"
   
   # --- MISO Quantification for Control Samples ---
   # Iterate through control BAM files and run MISO for each sample to quantify isoform usage (PSI values).
   for bam_file in "${CONTROL_BAM_DIR}"/*.bam; do
       sample_name=$(basename "${bam_file}" .bam)
       echo "Running MISO quantification for control sample: ${sample_name}"
       # The --read-len parameter should match the actual read length of your RNA-Seq data.
       # Adjust other parameters like --num-reads-per-group if needed.
       miso --run "${MISO_ANNOTATIONS}" "${bam_file}" --output-dir "${OUTPUT_DIR_CONTROL_QUANT}/${sample_name}" --read-len 50
   done
   
   # --- MISO Quantification for Diabetic Samples ---
   # Iterate through diabetic BAM files and run MISO for each sample.
   for bam_file in "${DIABETIC_BAM_DIR}"/*.bam; do
       sample_name=$(basename "${bam_file}" .bam)
       echo "Running MISO quantification for diabetic sample: ${sample_name}"
       # The --read-len parameter should match the actual read length of your RNA-Seq data.
       miso --run "${MISO_ANNOTATIONS}" "${bam_file}" --output-dir "${OUTPUT_DIR_DIABETIC_QUANT}/${sample_name}" --read-len 50
   done
   
   # --- MISO Comparison between Control and Diabetic Samples ---
   # This step compares the MISO output directories from the two groups
   # to identify differential splicing events (e.g., changes in skipped exon usage).
   # It computes Bayes factors and delta PSI values.
   echo "Running MISO comparison between control and diabetic samples"
   compare_miso "${OUTPUT_DIR_CONTROL_QUANT}" "${OUTPUT_DIR_DIABETIC_QUANT}" "${OUTPUT_DIR_COMPARISON}"
   

   Copy
9. 9

   The proccessed data format is tab deliniated .txt files.

   N/A (Inferred with models/gemini-2.5-flash) vN/A

   $ Bash example

   # The step description "The proccessed data format is tab deliniated .txt files" describes the expected output format of a previous processing step, rather than a processing step itself.
   # As no specific tool, process, or parameters are mentioned, a concrete bash command for a bioinformatics tool cannot be generated.
   # This description typically serves as metadata for the output of a preceding step (e.g., peak calling, quantification, differential expression results).
   #
   # Example of how one might inspect such a file (not a processing command):
   # head -n 5 processed_data.txt
   # # To check the number of columns (assuming tab-delimited):
   # # head -n 1 processed_data.txt | awk -F'\t' '{print NF}'
   

   Copy

Tools Used