zrcjessica / make_seurat

Run cellranger on snRNA-seq fastq files and create Seurat objecft

make_seurat

This pipeline runs cellranger-count with a custom reference genome on FASTQ files for scRNA-seq experiments and generates Seurat objects (.rds) from the resulting count data with corresponding QC plots.

This pipeline was designed for analyzing snRNA-seq from outbred rats classified by addiction index following self-administration of oxycodone or cocaine, in addition to rats that were never exposed to any drug (naive).

Usage

To run this pipeline, fill out the fields in the config file with your respective paths to the specified files and directories.

config.yml

out

Specify the relative path to the directory where output files will be written

data

Specify path where you want outputs of cellranger count to be written - can specify the same directory as out

fastqs_ref

This specifies a .tsv file with four columns:

RFID	Treatment	Addiction index	FASTQ prefix
Should correspond to sample names	e.g. cocaine, oxycodone, naive	e.g. high, low, none	Sample prefix for all FASTQ files for this sample; see 10x docs for details

The tsv file should keep the row of header names. Header names can be changed.

fastqs_dir

Path to directory containing subdirectories each named after a sample (RFID). Each subdirectory contains all the FASTQ files for each sample.

e.g.

fastqs_dir
├── SampleA
│   ├── SampleA_1_S41_L002_R1_001.fastq.gz
│   ├── SampleA_1_S41_L002_R2_001.fastq.gz
│   ├── SampleA_2_S42_L002_R1_001.fastq.gz
│   ├── SampleA_2_S42_L002_R2_001.fastq.gz
│   ├── SampleA_3_S43_L002_R1_001.fastq.gz
│   ├── SampleA_3_S43_L002_R2_001.fastq.gz
│   ├── SampleA_4_S44_L002_R1_001.fastq.gz
│   ├── SampleA_4_S44_L002_R2_001.fastq.gz
│   ├── SampleA_S2_L001_R1_001.fastq.gz
│   └── SampleA_S2_L001_R2_001.fastq.gz
├── SampleB
│   ├── SampleB_S4_L002_R1_001.fastq.gz
│   └── SampleB_S4_L002_R2_001.fastq.gz
├── SampleC
│   ├── SampleC_S3_L002_R1_001.fastq.gz
│   └── SampleC_S3_L002_R2_001.fastq.gz

ref_transcriptome

Path to 10x custom reference genome built with cellranger mkref for your organism.

cellranger count input params

Can specify expect_cells, chemistry, localmem, and localcores. See 10x Command-Line Argument Reference for details.

Seurat CreateSeuratObject() arguments

Can specify min.cells and min.features. See docs here.

samples

List of sample names corresponding to first column of tsv file referenced by fastqs_ref.