Frequently asked questions¶
Can this program work with dual barcodes / indexes?
Yes, but not directly. Because of the large amount of dual (or more) indexing approaches, the user interface would become incomprehensible. This is why we have decided to support only the basic cases. In order to support an arbitrary amount of barcodes see the Multiple barcodes section.
Can you add support for removing barcodes after demultiplexing?
We used to have this type of functionality (selecting parts of a read) in a previous version, but because of the large number of complicated barcoding schemas (multiple barcodes in one read, barcodes in multiple reads, etc.), we found that this interface was not flexible enough. Instead, we recommend to use a more generic tool for post processing the demultiplexed files, The Fastools
selectcommand for example.
My sequencing run was pretty bad, can / should I increase the number of allowed mismatches?
It depends on which barcodes were used. Most barcode sets are designed to allow for single nucleotide read errors. When multiple errors occur, it may not be possible to uniquely assign a read to a barcode. You can use the Barcode
testcommand to see if your barcode set allows for multiple error correction.
I do not know which / how many barcodes were used. How can I demultiplex my file?
The best thing to do is to contact your sequencing provider and ask which barcodes were used. If this is not possible for some reason, you may want to use the
guesssubcommand described in the Illumina FASTQ files section. If the barcodes are in the read instead of the header, you may want to use a tool like FastQC to find overrepresented sequences. These may be the barcodes you are looking for.
I get the message “error: invalid barcodes file format”. What is wrong?
The columns in
barcodes.tsvshould be separated by spaces or tabs, using other delimiters will result in this error message. Also note that thedemuxsubcommand only allows for one barcode, while thematchcommand can work with multiple barcodes.