Illumina paired-end sbRNA-seq pipeline

Nextflow pipeline to process already demultiplexed Illumina paired-end FASTQ files from multiple bacterial samples into a gene $\times$ cell count table.

Processing steps

For each set of barcodes:

Generate all combinations to make a cell barcode whitelist.
For each cell barcode, generate all 1-error variants for parsing by umi-tools.

For each reference genome:

Download the genome FASTA and GFF.

For each sample:

Trim reads to adapters using cutadapt. Reads from sbRNA-seq will be flanked by sequences containing cell barcodes and UMIs, so this step trims any extra sequences either side of these flanking sequences.
Extract cell barcodes and UMIs using umi-tools.
Map to genome FASTA using bowtie2.
Deduplicate mapped reads using umitools.
Count deduplicated reads per gene using featureCounts.
Count deduplicated reads per gene per cell using umi-tools.
Plot:
- Histogram of UMIs per cell per gene
- Histogram of unique genes per cell
- Histogram of unique cells per gene
- Scatter of UMI count vs gene count
- Scatter of UMI count vs cell count
Use scanpy to filter cells to have > 20 unique genes, and filter genes to be found in > 0 cells, run basic QC, Leiden cluster and UMAP embed with default settings, and save resulting plots and AnnData object as .h5ad for downstream analysis.

Downstream analyses [work in progress, not yet implemented]

De novo transcriptome assembly with Trinity.
Isoform analysis with RSEM.
Plotting and visualisation.

Other steps

Get FASTQ quality metrics with fastqc.
Compile the logs of processing steps into an HTML report with multiqc.

Requirements

Software

You need to have Nextflow and conda installed on your system.

First time using Nextflow?

If it's your first time using Nextflow on your system, you can install it using conda:

conda install -c bioconda nextflow

You may need to set the NXF_HOME environment variable. For example,

mkdir -p ~/.nextflow
export NXF_HOME=~/.nextflow

To make this a permanent change, you can do something like the following:

mkdir -p ~/.nextflow
echo "export NXF_HOME=~/.nextflow" >> ~/.bash_profile
source ~/.bash_profile

Quick start

Make a sample sheet (see below) and, optionally, a nextflow.config file in the directory where you want the pipeline to run. Then run Nextflow.

nextflow run scbirlab/nf-sbrnaseq

Each time you run the pipeline after the first time, Nextflow will use a locally-cached version which will not be automatically updated. If you want to ensure that you're running a version of the pipeline, use the -r <version> flag. For example,

nextflow run scbirlab/nf-sbrnaseq -r v0.0.4

A list of versions can be found by running nextflow info scbirlab/nf-sbrnaseq.

For help, use nextflow run scbirlab/nf-sbrnaseq --help.

The first time you run the pipeline on your system, the software dependencies in environment.yml will be installed. This may take several minutes.

Inputs

The following parameters are required:

sample_sheet: path to a CSV containing sample IDs matched with FASTQ filenames, genome information, and adapter sequences.

If you're using local FASTQ data (the default, from_sra = false), you also need to supply the parameter:

fastq_dir: path to directory containing the FASTQ files

The following parameters have default values can be overridden if necessary.

from_sra = false: whether to fetch FASTQ files from an SRA Run ID instead of loading local files. In this case, your sample sheet needs a column headed Run to indicate the Run ID.
allow_cell_errors = true: Whether to allow 1 error when matching cell barcodes in the whitelist. Otherwise only exact matches are allowed. This is ignored when the total length of the cell barcode is too long.
inputs = "inputs": path to directory containing files referenced in the sample_sheet, such as lists of guide RNAs.
output = "outputs": path to directory to put output files
trim_qual = 5 : For cutadapt, the minimum Phred score for trimming 3' calls
min_length = "9:38" : For cutadapt, the minimum trimmed length of a read. Shorter reads will be discarded
strand = 1 : For featureCounts, the strandedness of RNA-seq. 1 for forward, 2 for reverse.
ann_type = 'gene' : For featureCounts, features from GFF column 3 to use for counting
label = 'Name' : For featureCounts, one or more (comma-separated) fields from column 9 of GFF for labeling counts

The parameters can be provided either in the nextflow.config file or on the nextflow run command.

Here is an example of the nextflow.config file:

params {
   
    sample_sheet = "/path/to/sample_sheet.csv"
    fastq_dir = "/path/to/fastqs"

    trim_qual = 15
}

Alternatively, you can provide these on the command line:

nextflow run scbirlab/nf-sbrnaseq \
    --sample_sheet /path/to/sample_sheet.csv \
    --fastq_dir /path/to/fastqs \
    --trim_qual 15

Sample sheet

The sample sheet is a CSV file providing information about each sample: which FASTQ files belong to it, the reference genome accession number, adapters to be trimmed off, and the UMI and cell barcode scheme for each sample.

The file must have a header with the column names below, and one line per sample to be processed.

sample_id: the unique name of the sample
genome_id: The NCBI assembly accession number for the organism that the guide RNAs are targeting. This number starts with "GCF_" or "GCA_".

If using local files (from_sra = false):

fastq_pattern: the search glob to find FASTQ files for each sample in fastq_dir. The pipleine will look for files matching <fastq_dir>/*<fastq_pattern>*, and should match only two files, corresponding to paired reads.

If using SRA files (from_sra = false):

Run: the SRA run ID of this sample. When using SRA-derived FASTQ files, the Illumina indices will be prepended to the read, so you will have to account for this in your adapter_read* columns and umi_read* columns

Other required columns:

bc1, bc2, bc3: Files containg barcodes corresponding to <cell_1>, <cell_2>, and <cell_3> in umi_read* columns.
adapter_read1_3prime: the 3' adapter on the forward read to trim to in cutadapt format. The adapter itself and sequences downstream will be removed.
adapter_read2_3prime: the 3' adapter on the reverse read to trim to in cutadapt format. The adapter itself and sequences downstream will be removed.
adapter_read1_5prime: the 5' adapter on the forward read to trim to in cutadapt format. Sequences upstream will be removed, but the adapters themselves will be retained.
adapter_read2_5prime: the 5' adapter on the reverse read to trim to in cutadapt format. Sequences upstream will be removed, but the adapters themselves will be retained.
umi_read1: the cell barcode and UMI pattern in umitools regex format for the forward read
umi_read2: the cell barcode and UMI pattern in umitools regex format for the reverse read

Here is an example of the sample sheet:

sample_id	fastq_pattern	genome_id	bc1	bc2	bc3	adapter_read1_3prime	adapter_read2_3prime	adapter_read1_5prime	adapter_read2_5prime	umi_read1	umi_read2
EcoHX1	G5512A22_R	GCF_904425475.1	bc1.csv	bc2.csv	bc3.csv	CAGN{6}G{3}	N{7}N{8}TTATTATA	TATAATAAN{8}N{7}	C{3}N{6}CTG	^(?P<discard_1>.{3})(?P<cell_1>.{6}).*	^(?P<discard_2>.{8})(?P<cell_2>.{8})(?P<umi_1>.{7}).*
EcoHX2	G5512A23_R	GCF_904425475.1	bc1.csv	bc2.csv	bc3.csv	CAGN{6}G{3}	N{7}N{8}TTATTATA	TATAATAAN{8}N{7}	C{3}N{6}CTG	^(?P<discard_1>.{3})(?P<cell_1>.{6}).*	^(?P<discard_2>.{8})(?P<cell_2>.{8})(?P<umi_1>.{7}).*

And here is an example barcode file:

Barcode ID	Sequence
barcode-01	ATCGATCC
barcode-02	TAGCTAGG

It must have a header row and two columns, with the second column being the barcode sequence.

Outputs

Outputs are saved in the same directory as sample_sheet. They are organised under three directories:

processed: FASTQ files and logs resulting from trimming and UMI extraction
counts: tables and BAM files corresponding to cell $\times$ gene counts
multiqc: HTML report on processing steps

Issues, problems, suggestions

Add to the issue tracker.

Further help

Here are the help pages of the software used by this pipeline.