pacbio-kinnex-nf

Background

pacbio-kinnex-nf is an implementation of the PacBio Single-Cell Iso-Seq pipeline in Nextflow. It takes Circular Consensus Sequencing (CCS) reads, the output of the SMRTLINK software, and processes them using the CLI Workflow from isoseq.how.

Usage

Clone the repository

git clone https://github.com/kevinpryan/nf-hlamajority.git

Run the pipeline on test data with:

nextflow run main.nf \
       -profile test,<singularity/docker/...>

The PBBM2_ALIGN step is computationally intensive, even when running on the test dataset. The test profile requests 8 CPUs and 64 GB of RAM for this step.

To run on your own data, place all your CCS BAMs in the same directory and run:

nextflow run main.nf \
       -profile <singularity/docker/...> \
       -outdir <outdir_name> \
       --bams "/path/to/bam/dir/*.bam"

Expected output from running test profile:

├── multiqc
│   ├── multiqc_data
│   │   ├── isoseq_refine_boxplot_fivelen.txt
│   │   ├── isoseq_refine_boxplot_insertlen.txt
│   │   ├── isoseq_refine_boxplot_polyAlen.txt
│   │   ├── isoseq_refine_boxplot_threelen.txt
│   │   ├── llms-full.txt
│   │   ├── multiqc_citations.txt
│   │   ├── multiqc_data.json
│   │   ├── multiqc_isoseq_refine_csv.txt
│   │   ├── multiqc_isoseq_refine_json.txt
│   │   ├── multiqc.log
│   │   ├── multiqc.parquet
│   │   └── multiqc_sources.txt
│   └── multiqc_report.html
├── pigeon_seurat
│   └── ccs_pigeon_seurat
│       ├── ccs.annotated.info.csv
│       ├── ccs.info.csv
│       ├── genes_seurat
│       │   ├── barcodes.tsv
│       │   ├── genes.tsv
│       │   └── matrix.mtx
│       └── isoforms_seurat
│           ├── barcodes.tsv
│           ├── genes.tsv
│           └── matrix.mtx
├── pipeline_info
│   ├── execution_report.html
│   ├── execution_timeline.html
│   └── execution_trace.txt
└── reference_cache
    └── reference.gtf

Intermediate files can be found in the relevant process work directory.

Pipeline description

The steps are as follows:

Primer removal and identification of barcodes (LIMA)

Inputs:

• CCS BAM
• Primers fasta (default is primers from 10x 3' kit)

primers.fasta

>5p
AAGCAGTGGTATCAACGCAGAGTACATGGG
>3p
AGATCGGAAGAGCGTCGTGTAG

Outputs:

• BAM with primers removed, correctly oriented sequences. Name of primer sequences will be inserted into BAM name (e.g. for test data, BAM output name is ccs.lima.output.5p--3p.bam)
• BAM index (*.bam.pbi)
• Consensus read set XML (*.consensusreadset.xml)
• Counts (*.output.lima.counts)
• Summary (*.output.lima.summary)

Tag reads with UMIs and barcode information (ISOSEQ_TAG)

Clip UMIs and cell barcodes from reads and associate with the reads for deduplication.

Inputs:

• LIMA.out.bam
• UMI design (default T-12U-16B, customise by passing --design my-design when running pipeline)

Outputs:

• Tagged BAM ("*.flt.bam")
• BAM index ("*.flt.bam.pbi")

Refine reads (ISOSEQ_REFINE)

Trims poly(A) tails and removes unintended concatemer

Inputs:

• ISOSEQ_TAG.out.bam
• Primers fasta

Outputs:

• Full-length non-concatemer reads ("*.fltnc.bam")
• Consensus read set XML ("*.fltnc.consensusreadset.xml")
• Summary report json, used in MULTIQC ("*.fltnc.filter_summary.report.json")
• Summary csv, used in MULTIQC ("*.fltnc.report.csv")

Correct cell barcode errors and identify real cells (ISOSEQ_CORRECT)

Identify and correct errors in cell barcodes. Requires a cell barcode whitelist, which can likely be found here.

Inputs:

• ISOSEQ_REFINE.out.bam
• Barcodes, default: 3M-february-2018-REVERSE-COMPLEMENTED.txt.gz

Outputs:

• Corrected BAM (*.corrected.bam)
• Corrected BAM index (*.corrected.bam.pbi)
• Intermediate BAM index (*.corrected_intermediate.bam.pbi)
• Corrected report json (*.corrected.report.json)