GSE240325 Processing Pipeline

Publication

Nature communications (2024) — PMID 39152130

Dataset

An in situ method for identification of transcriptome-wide protein-RNA interactions in cells [in_situ_STAMP - Long-Read]"

1. 1

   Demultiplex primers with lima (v2.6.0)

   lima v2.6.0

   $ Bash example

   # Install lima (e.g., via Bioconda)
   # conda install -c bioconda lima=2.6.0
   
   # Example usage of lima for demultiplexing primers
   # Replace 'input_reads.fastq' with your actual input sequencing reads file (e.g., PacBio CCS reads).
   # Replace 'primers.fasta' with your FASTA file containing barcode sequences.
   # Replace 'output_demultiplexed_reads' with your desired output prefix.
   # The --ccs flag is commonly used if the input reads are PacBio CCS reads.
   # Adjust other parameters like --min-score, --num-threads, --split-bam-by-barcode as needed based on your data and requirements.
   lima input_reads.fastq primers.fasta output_demultiplexed_reads --ccs --peek-guess --min-score 20 --num-threads 8

   Copy
2. 2

   Refine reads with Isoseq3 refine (v3.8.0)

   IsoSeq v3.8.0 GitHub

   $ Bash example

   # Install IsoSeq3 via Bioconda
   # conda create -n isoseq3_env python=3.8
   # conda activate isoseq3_env
   # conda install -c bioconda isoseq3=3.8.0
   
   # Example command for refining reads with Isoseq3 (v3.8.0)
   # This step typically follows 'isoseq3 cluster' and requires its output.
   # Replace 'flnc_reads.bam', 'primer_sequences.fasta', 'cluster_report.csv', and 'refined_reads.bam' with actual file paths.
   
   # Input files:
   # --flnc: Full-length non-chimeric (FLNC) reads BAM file (output from 'lima' or 'isoseq3 cluster' if not using 'lima')
   # --primer: FASTA file containing primer sequences used for library preparation
   # --cluster-report: CSV report from 'isoseq3 cluster' containing cluster information
   
   # Output files:
   # --output: Refined FLNC reads BAM file
   # --gff: GFF file containing refined transcript annotations
   
   isoseq3 refine \
       --flnc flnc_reads.bam \
       --primer primer_sequences.fasta \
       --cluster-report cluster_report.csv \
       --output refined_reads.bam \
       --gff refined_reads.gff

   Copy View on GitHub
3. 3

   Align reads using pbmm2 (v1.9.0)

   pbmm2 v1.9.0 GitHub

   $ Bash example

   # Install pbmm2 if not already installed
   # conda install -c bioconda pbmm2=1.9.0
   
   # Define input/output files and reference genome
   # Replace 'reads.fastq' with your actual input reads file (e.g., .fastq, .fasta, .bam)
   READS_FILE="reads.fastq"
   
   # Replace 'reference.fasta' with the path to your reference genome file (e.g., GRCh38.fasta)
   # pbmm2 can also use an .mmi index file if pre-built (e.g., reference.mmi)
   REFERENCE_GENOME="reference.fasta"
   
   # Define the output BAM file name
   OUTPUT_BAM="aligned_reads.bam"
   
   # Align reads using pbmm2
   # This command performs alignment and outputs a sorted BAM file by default.
   pbmm2 align "${REFERENCE_GENOME}" "${READS_FILE}" "${OUTPUT_BAM}"

   Copy View on GitHub
4. 4

   Cluster reads with isoseq3 cluster (v3.8.0)

   IsoSeq v3.8.0 GitHub

   $ Bash example

   # Install isoseq3 via conda if not already installed
   # conda install -c bioconda -c conda-forge pbbioconda
   
   # Cluster reads using isoseq3 cluster
   # Replace input.fasta with your actual input FASTA/FASTQ file containing polished reads
   # Replace output_prefix with your desired output file prefix
   isoseq3 cluster --input input.fasta --output output_prefix

   Copy View on GitHub
5. 5

   Filter for primary mapped reads with custom script

   samtools (Inferred with models/gemini-2.5-flash) v1.10 (Inferred with models/gemini-2.5-flash) GitHub

   $ Bash example

   # Install samtools if not available
   # conda install -c bioconda samtools
   
   # Define input and output file names
   INPUT_BAM="input.bam"
   OUTPUT_BAM="primary_mapped.bam"
   
   # This is a custom script to filter for primary mapped reads.
   # It uses samtools view to exclude secondary and supplementary alignments.
   # Additional filtering logic (e.g., mapping quality, read length, specific flags)
   # could be added here based on specific assay requirements.
   
   # Filter for primary mapped reads (flag -F 256 excludes secondary and supplementary alignments)
   # -b: output in BAM format
   samtools view -b -F 256 "${INPUT_BAM}" > "${OUTPUT_BAM}"
   
   # Index the filtered BAM file for downstream processing
   samtools index "${OUTPUT_BAM}"

   Copy View on GitHub
6. 6

   Identify gene editing using custom script

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

   $ Bash example

   # Placeholder for a custom gene editing identification script.
   # This script would typically take aligned reads (BAM/CRAM) or variant calls (VCF)
   # and a reference genome to identify specific gene edits.
   
   # Define input and output files (placeholders)
   INPUT_FILE="input_variants.vcf" # Or aligned_reads.bam
   REFERENCE_GENOME="/path/to/reference/hg38.fa"
   OUTPUT_REPORT="gene_editing_report.txt"
   
   # Example execution command for a hypothetical custom Python script
   # Replace 'custom_gene_editing_script.py' with the actual script name
   # and adjust parameters as needed for the specific custom script.
   python custom_gene_editing_script.py \
     --input "${INPUT_FILE}" \
     --reference "${REFERENCE_GENOME}" \
     --output "${OUTPUT_REPORT}"
   

   Copy
7. 7

   remove edits found in annotated SNP positions using custom script

   filter_snps.py (Inferred with models/gemini-2.5-flash) vN/A GitHub

   $ Bash example

   # Install dependencies if not already available
   # conda install -c bioconda pysam pybedtools
   
   # Assuming 'input_edits.bed' is the file containing detected edits (e.g., from variant calling or RNA editing detection)
   # Assuming '/path/to/human_grch38_known_snps.vcf.gz' is a VCF file of known SNPs for the human GRCh38 reference genome (e.g., from dbSNP or gnomAD)
   
   # Execute the custom script to remove edits found in annotated SNP positions
   python filter_snps.py \
       --input-bed input_edits.bed \
       --output-bed filtered_edits.bed \
       --vcf /path/to/human_grch38_known_snps.vcf.gz

   Copy View on GitHub