A toolbox for correcting barcodes in 10X linked-read sequencing data.
If everything is setup correctly, this will create the binary 'bcctools'.
The only input needed for barcode correction is a pair of barcoded FASTQ files generated on the 10X Chromium platform. Optionally, you can specify a barcode whitelist file.
The program consists of several commands, which are listed when running
./bcctools --help
For a short description of each command and an overview of arguments and options, you can run
./bcctools <COMMAND> --help
If you need the output to be sorted and/or converted to SAM, BAM, or (gzipped) FASTQ format, you can run the provided bash script. For a short description of options and arguments of this script run
./scripts/run_bcctools -h
./bcctools whitelist [OPTIONS] <FASTQ 1 file>
Creates a barcode whitelist based on barcode occurence in the data. Creating a whitelist from your data is recommended (rather than using the 10X whitelist) to reduce the number of alternatives during correction and prevents false corrections.
./bcctools index [OPTIONS] <whitelist file>
Creates a barcode index from the given barcode whitelist and writes it to disk. This command is optional as the index can be created on the fly in the 'correct' command.
./bcctools correct [OPTIONS] <whitelist file> <FASTQ 1 file> <FASTQ 2 file>
Corrects barcodes of the given barcoded read pair data using the specified barcode whitelist. A barcode index is computed on the fly unless index files are present for the specified barcode whitelist. The output is a tab-separated file holding one read pair per line as decribed below.
./bcctools stats [OPTIONS] <Corrected (gzipped) FASTQ 1 file>
./bcctools stats [OPTIONS] <Corrected SAM/BAM file>
./bcctools stats [OPTIONS] <Corrected TSV file>
Computes the number of read pairs with whitelisted, corrected and unrecognized barcodes, a barcode occurrence histogram and counts quality values of corrected barcode positions.
mkdir bcctools_example && cd bcctools_example/
ln -s /path/to/first.fq.gz
ln -s /path/to/second.fq.gz
./bcctools whitelist -o whitelist.txt first.fq.gz
./bcctools correct whitelist.txt first.fq.gz second.fq.gz > corrected.tsv
Using the bash script to create a BAM file sorted by the corrected barcode sequence:
./script/run_bcctools -f bam first.fq.gz second.fq.gz
The output format of the correct command is a simple tab-separated format, where each read pair and its barcode information is given on a single line. The fields are as follows:
Field | Description |
---|---|
READ NAME | The read or query name taken from the FASTQ file and cropped at the first whitespace. |
CORRECTED BARCODE | A comma separated list of possible barcode corrections. If the raw barcode is whitelisted, the value of this field is identical to the RAW BARCODE field. An asterisk '*' indicates that the barcode is not whitelisted and correction was unsuccessful. |
RAW BARCODE | The first 16 base pairs of the first read in the read pair. |
7-MER SPACER | The seven base pairs following the first 16 base pairs of the first read in the read pair. |
TRIMMED FIRST READ | The remaining base pairs of the first read in the read pair after trimming the barcode and 7-mer spacer sequence. |
SECOND READ | The second read sequence. |
BARCODE QUALITY STRING | The first 16 values of the quality string of the first read in the read pair. |
7-MER SPACER QUALITY STRING | The seven values following the first 16 values of the quality string of the first read in the read pair. |
TRIMMED FIRST READ QUALITY STRING | The remaining quality string after trimming the barcode and 7-mer spacer quality values. |
SECOND READ QUALITY STRING | The quality string of the second read in the read pair. |
For questions and comments contact birte.kehr [at] ukr.de or create an issue.