For the complete documentation index, see llms.txt. This page is also available as Markdown.

Why are underscores in sample names changed to dashes in the FASTQ file names?

Any underscores in Sample ID and Sample Names are converted to dashes to preserve sample naming conventions required for downstream analysis. Illumina FASTQs use the following format: SampleName_SampleNumber_Lane_Read_FlowCellIndex.fastq.gz

Local Run Manager, MiSeq Reporter, bcl2fastq, and BCL Convert when used through BaseSpace Sequence Hub (BSSH) will automatically make this change. However, locally-run BCL Convert, locally-run bcl2fastq, and onboard analysis on the NovaSeq X and NextSeq 1000/2000 will keep underscores exactly as provided in the input.

  • Because the underscore-to-dash conversion can cause two distinct sample names to resolve to the same FASTQ filename, do not rely on underscores/dashes as the sole distinguishing characters between sample names. In instances when the underscore-to-dash conversion leads to samples with non-unique names, this can lead to analysis prematurely terminating or overwriting of data, depending on the software.

Additionally, note that FASTQs containing underscores in the SampleName field cannot be uploaded to BSSH; any underscores in sample names must be replaced with dashes or removed prior to upload. This requirement is described in the article Requirements for uploading FASTQ Files to BaseSpace via web interface.

For any feedback or questions regarding this article (Illumina Knowledge Article #3636), contact Illumina Technical Support techsupport@illumina.com.

Last updated

Was this helpful?