Third-generation single cell sequencing (Pacbio) analysis pipeline.

This pipeline is developed to analysis third-generation sequencing data produced by Pacbio platform. It takes CCS bam file as input file and generate output files which can be used for downstream data analysis as well as a summary QC statistic file and two html report of transcriptome and alternative splicing atlas.

The two main functions of this pipeline are to analyze transcript characteristics and alternative splicing characteristics of Pacbio dataset. The pipeline consists of several steps and the first 6 steps were identical pre-processing steps, refer to ISOseq guidline:

• ccs HiFi reads which produced using circular consensus sequencing (CCS) mode on PacBio long-read systems are took as input file and this step generate the directory structure for CCS bam file.

• remove_primer lima is used to remove the 5' and 3' cDNA primers.

• pattern_detection UMIs and cell barcodes are clipped from the reads and associated with the reads for later deduplication, refer to isoseq3 tag.

• flc isoseq3 refine is used to remove polyA tail and artificial concatemers.

• split_linker Remove linker sequence in barcode.

• dedup This step performs PCR deduplicatation via clustering by UMI and cell barcodes, refer to isoseq3 dedup.

To analysis transcript characteristics, refer to celescope:

• featurecount Assigning uniquely mapped reads to genomic features with FeatureCounts.

• count Distinguish cell barcodes from background barcodes and Generate expression matrix.

• run_seurat Cell clustering with Seurat and Calculate the marker gene of each cluster.

To analysis alternative splicing characteristics:

• run_isoform Applying Cupcake to collapse the mapped dedup-ed reads into unique transcripts and SQANTI3 to annotate each unique transcript against an annotation.

v1.1

It is recommended to use mamba, download and install mamba.

wget https://github.com/conda-forge/miniforge/releases/latest/download/Mambaforge-Linux-x86_64.sh
bash Mambaforge-Linux-x86_64.sh

git clone https://github.com/singleron-RD/Pacbio_Analysis.git

cd Pacbio_Analysis

mamba create -n pacbio_isoseq -y --file conda_pkgs.txt

mamba activate pacbio_isoseq

cd Pacbio_Analysis/tools/cDNA_Cupcake

python setup.py build

python setup.py install

export PYTHONPATH=$PYTHONPATH:<path_to_Cupcake>/sequence/

python src/run_pacbio.py --ccs_bam input_bam_file -- genomeDir genome_directory --sample sample_name <optional parameters>

--ccs_bam CCS bam file, required

--genomeDir Reference genome directory, required

--sample Sample name/Output file prefix, required

--pacbio_source_path Absolute path to Pacbio_analysis

--steps set specific steps to run, "all" to run all the steps, optional steps include: ccs, remove_primer, pattern_detection, flc, split_linker, dedup, featurecount, count, run_seurat, run_isoform, required

--outdir Output file directory, default current

--primer_fasta Primer sequence file, fasta fomat, optional

--bu_pattern Barcord and UMI pattern expression, default T-12U-57B

--blastn Blastn tool path

--bclist Barcode sequence file, used in linker_split step

--report Whether to generate summary statistical report, default True

To analysis isoform only (Recommend):

cd test
mkdir test_res_isoform
conda activate pacbio_v1
python src/run_pacbio.py 
    --ccs_bam  <path>/test.bam
    --genomeDir genome_directory
    --pacbio_source_path <path>/Pacbio_Analysis
    --sample test_sample
    --outdir test_res_isoform
    --steps ccs,remove_primer,pattern_detection,flc,split_linker,dedup,run_isoform

To run all the steps (Calculate expression matrix with Celescope):

cd test
mkdir test_res_all
conda activate pacbio_v1
python src/run_pacbio.py 
    --ccs_bam  <path>/test.bam
    --genomeDir genome_directory
    --pacbio_source_path <path>/Pacbio_Analysis
    --sample test_sample
    --outdir test_res_isoform
    --steps all

If all the steps completed, we'll get:

outdir
|--01.ccs
|--02.remove.primers
|--03.pattern.bc.umi
|--04.flc
|--05.split.linker
|--06.dedup
|--07.featurecount
|--08.count
|--09.seurat
|--10.isoform
|--11.annotation
|--sample.Pacbio_qc.stat.txt
|--sample_report.html
|--sample.SQANTI3_report.html

Provide the absolute path of ccs bam file to avoid unexpected errors.