Repository of stripts for processing PacBio IsoSeq data

Data structure

Datasets are organised by processing stage and then into folders by project name.

project
│   README.md
│   file001.txt    
│
└───RawData
   │
   └───PacBio

   └───RNASeq
       └───Project1
           │   sample1_R1.fastq.gz
           │   sample1_R2.fastq.gz
           │   ...       
           └───fastqc
               | sample1_fastqc.zip
               | sample1_fastqc.html
               | ...
               | multiqc_report.html
               └───multiqc_data 
           └───trimmed
               | sample1_R1_val_1.fq.gz
               | sample1_R2_val_2.fq.gz
               | sample1_R1_val_1_fastqc.zip
               | sample1_R2_val_2_fastqc.zip 
               | sample1_R1_val_1_fastqc.html
               | sample1_R2_val_2_fastqc.html            
               | sample1_R1_val_1.fq.gz_trimming_report.txt
               | sample1_R1_val_1.fq.gz_trimming_report.txt               
               | ...          
   └───Aligned
    │
    └───RNASeq
       └───Project1
           | sample1_Aligned.sortedByCoord.out.bam
           | sample1_Log.final.out
           | sample1_Log.progress.out
           | sample1_Log.out
           | sample1_SJ.out.tab           
           | ...       
           | multiqc_report.html
           └───rnaseqc             
           └───multiqc_data           
                     
└───GeneCounts
  │
  └───Project1
    └───RSEM
      └───GENCODE
        │   sample1.isoforms.results
        │   sample1.genes.results
      └───BrainTAMAPost
        │   sample1.isoforms.results
        │   sample1.genes.results
└───Scripts
    │   file021.txt
    │   file022.txt    

Details for running the isoseq pipeline

Run the following commands from the github repository main folder

SMRT cell pre-processing

1. sbatch Scripts/JobSubmission/batchProcessIsoSeqData.sh searchs for all files with sufix .subreads.bam then executes the following scripts for each sample:

   • processIsoSeqSMRTcells.sh runs the isoseq3 pipeline including ccs, lima, refine, cluster and polish steps.
   • alignIsoSeqSMRTcells.sh aligns the processed data using minimap2
   • filterIsoSeqSMRTcells.sh performs filtering on the on the aligned transcripts using cupcake scripts
   • alignShortReadSMRTcells.sh aligns short read data to gff for use in sqanti qc
   • sqanti3SMRTcells.sh QCs SMRT-cell level transcriptomes prior to merging

2. smrt cell level qc summary

create master transcriptome with TALON

1. sbatch Scripts/IsoSeqPipeline/createMasterTranscriptomeFromSMRTcells.sh use talon to mark reads for internal priming and merge smrt cell transcriptomes into single master transcriptomes. Compare to GENCODE for annotation of known isoforms

2. sbatch Scripts/IsoSeqPipeline/.sh use talon to annotate master transcriptome with

3. characterise transcriptome

create master transcriptome with TAMA

1. sbatch Scripts/JobSubmission/batchProcessTamaPipeline.sh preprocesses each smrt cell for use with tama by executing

   • alignFLNC.sh aligns output of refine step (i.e. skips cluster and polishing steps)
   • tamaCollapse.sh uses tama to collapse isoforms

2. sbatch Scripts/JobSubmission/createMasterTranscriptomeWithTama.sh use tama to merge smrt cells and perform some filtering

3. sbatch Scripts/JobSubmission/batchAlignSRTama.sh align short read data to merged transcriptome for input to sqanti QC

4. Scripts/JobSubmission/qcTamaTranscriptome.sh use sqanti to characterise and filter isoforms

RNA-Seq data pipeline

1. sbatch --array= Scripts/JobSubmission/runRNASeqAlignment.sh

   • Scripts/RNASeq/alignShortReadData.sh: fastqc, trimGalore, align with STAR to GENCODE
   • Scripts/RNASeq/rsemGENCODE.sh: quantify GENCODE transcripts with RSEM
   • Scripts/RNASeq/rnaseqQC.sh: RNASEQ-QC

2. sbatch --array= Scripts/JobSubmission/runAlignmentBrainTranscriptome.sh

   • Scripts/RNASeq/rsemBrainTranscriptome.sh: quantify Brain Transcriptome with RSEM

3. sbatch JobSubmission/featureCountsGENCODE.sh

   • uses featureCounts to calculate exon and gene level counts for GENCODE transcriptome

Differential Expression analyses

1. Rscript Scripts/DiffSplicing/edgeRExpression.r