Demultiplex single-cells into individual files

Let’s say you have single-cell sequencing reads, such as those produced by 10x Genomics, where the 16 bp barcodes (used to label individual cells) are at the beginning of the first read file (R1.fastq) but you want to put each individual cell into its own file. This can be done using splitcode’s demultiplexing capabilities (see the page: Demultiplexing for a more detailed description of splitcode’s demultiplexing capabilities).

Here, we’ll create the following config.txt file:

config.txt
 tags                 locations  distances
 GACGTAGCGCCCCCCA     0:0:16     1
 TAGCTACCTTGTAACA     0:0:16     1
 CCACCGAATAGGAACC     0:0:16     1
 TTGTAGCTGAGTAGTA     0:0:16     1
 CCTTCCTTAAACTTAC     0:0:16     1
 CCTTCCTTAAACTTAG     0:0:16     1

We showcase a series of 16 bp barcodes and specify that it occurs at the beginning (i.e. first 16 bps) of the first sequencing read via 0:0:16 and a hamming distance error of 1 (one substitution) is allowed. Of course, single-cell technologies can have thousands of barcodes, all of which would be put into this file via this format, but we only show a few here for demonstration purposes.

Next, we’ll create the following keep.txt file for our demultiplexing strategy (the first column is the tag ID, which by default is the tag sequence itself, and the second column is the prefix of the output file name):

keep.txt
 GACGTAGCGCCCCCCA     1
 TAGCTACCTTGTAACA     2
 CCACCGAATAGGAACC     3
 TTGTAGCTGAGTAGTA     4
 CCTTCCTTAAACTTAC     5
 CCTTCCTTAAACTTAG     6

We then run the following:

splitcode --gzip --keep-r1-r2 --keep=keep.txt -c config.txt --nFastqs=2 --no-output --no-outb R1.fastq R2.fastq

We use --no-output because we don’t have our “typical output” (i.e. those specified by --pipe or -o; for all entries in select.txt where the second column is blank, those would go into the “typical output”). We use -no-outb because we don’t have any “final barcodes” (since --assign is not specified) so we shouldn’t allocate files for them.

The following six pairs of files will be generated (the _R1.fastq.gz files are the R1 files and the _R2.fastq.gz files are the R2 files since we specified --keep-r1-r2):

  • 1_R1.fastq.gz and 1_R2.fastq.gz

  • 2_R1.fastq.gz and 2_R2.fastq.gz

  • 3_R1.fastq.gz and 3_R2.fastq.gz

  • 4_R1.fastq.gz and 4_R2.fastq.gz

  • 5_R1.fastq.gz and 5_R2.fastq.gz

  • 6_R1.fastq.gz and 6_R2.fastq.gz

Each of these files will contain the barcodes associated with the respective files (e.g. reads with the barcode TTGTAGCTGAGTAGTA will be thrown into the files prefixed with 4, i.e. 4_R1.fastq.gz and 4_R2.fastq.gz).