Analysis launch FAQ for the Illumina 5 Base WGS and Enrichment kits
Which app is used for analysis of Illumina 5-Base DNA Prep and Illumina 5-Base DNA Prep with Enrichment libraries? Where is analysis supported (BaseSpace (BSSH), Illumina connected Analytics (ICA), local server?)
Analysis will be done in the BaseSpace apps listed below. With the DRAGEN 4.4.6 release, there is a dropdown menu labeled "Methylation for Illumina 5-Base DNA Prep", and checking the "Enable 5-Base Methylation-Aware Algorithms" will activate the proper settings for analysis of 5-Base libraries. The same named options are used in ICA apps as well.
DRAGEN Germline (gDNA or cfDNA)
DRAGEN Somatic (gDNA or cfDNA)
DRAGEN Enrichment (Enrichment libraries)
Data can be analyzed on BSSH, ICA, or local server. On-board secondary analysis is not supported, but FASTQs can be generated on-board through BCL Convert, and then used as input for the pipeline in BSSH, ICA, or local server.
Will customers be able to launch analyses with a BaseSpace Basic account, or will they need a Professional BaseSpace account? Will the app launch require iCredits?
The DRAGEN apps are paid apps that require iCredits to run.
Are there any plans for local analysis?
Yes, local analysis with a server is available in DRAGEN v4.4+. Illumina recommends using the most current DRAGEN (currently v4.4.6) for local analysis.
There are no plans for on-instrument analysis at this time, it must be run on a standalone DRAGEN server.
Consult the analysis information site for details on how to launch.
The Illumina DRAGEN pipelines are designed for the 5-Base library prep kits.
The suggested command-line in the reference manual does not mention anything about read trimming and/or UMIs? Should users enable any options related to these?
Using the sample sheet template and DRAGEN pipelines as instructed in the user guide should cover the vast majority of a user’s trimming needs. The sample sheet contains trimming setting applied in BCL Convert to remove adapter bases. The 5-base DRAGEN pipelines have soft clipping enabled by default to remove remaining artifacts.
gDNA samples, to trim an additional, custom amount, this can be done in the sample sheet by altering the OverrideCycle setting (eg, “N5” to ignore/trim 5 bases).
The sample sheet template also contains settings ensure proper UMI handling in BCL Convert. For additional details on how to enable UMI collapsing, see the appropriate DRAGEN recipes for the use case. UMI collapsing is only recommended for Enrichment datasets to ensure sufficient data yield.
Can data be analyzed with DRAGEN Methylation? Any caveats to results interpretation?
The DRAGEN Methylation app is not supported or recommended for use.
While DRAGEN Methylation can be run on 5-Base data, it is primarily intended for bisulfite converted DNA (where un-methylated C are converted to T), and not 5-Base libraries (where methylated C are converted to T). Running DRAGEN Methylation with 5-Base data will produce lower alignment and no variant calls. Use DRAGEN Germline, DRAGEN Somatic, or DRAGEN Enrichment after checking the box to “Enable 5-Base Methylation Mode” for Illumina 5-Base data.
Can "Oncoanalyser" be used for analysis?
Oncoanalyser is intended for DNA or RNA data only, not DNA methylation FASTQ/BAM. Illumina recommends customers use the Illumina DRAGEN pipeline designed for the 5-Base library prep kit
Is a regular DRAGEN reference genome hash table used for analysis or is a special 5-Base version required?
A new DRAGEN reference genome hash is required that was created with DRAGEN Reference Builder v4.4.4+ and with "Include Methylation Data in Reference" checked. For human, this reference has been added to the DRAGEN cloud apps and download site so no customization building should be needed.
Image of option from DRAGEN Reference Builder v4.4.4:

What settings are required to create a 5-Base compatible hash table for Map/Align?
The analysis is compatible with all references that are labeled as “methyl_cg” (which is everything in the DRAGEN v11 Reference folder on ICA).
Regarding reference hash table selection for 5-Base, would there be any reason to pick the graph reference over linear for the hg38 v11 hash table, ie, hg38-alt_masked.cnv.graph.hla.methyl_cg.rna-11-r5.0-1 vs hg38-alt_masked.cnv.hla.methyl_cg.methylated_combined.rna-11-r5.0-1 (from here****)? There is an additional subdirectory "methyl_converted/" along with "methyl_cg/" in the linear reference, but it might not be relevant to 5-Base. Is that correct? This question applies to both germline (with small variant calling) and somatic (tumor only) pipelines.
Use graph genome reference for germline analyses.
Use linear genome reference for somatic analysis (somatic VC, shouldn't impact mC).
methyl_cg subdirectory is used for 5-Base data to account for conversion in CpG-dense regions. methyl_converted subdirectory is specific to 3-base methods (BiS/EMS) and used by the DRAGEN Methylation pipeline.
For somatic variant calling, does Illumina have a recommended file for systematic noise?
The systematic noise file Illumina recommends using (WGS_hg38_v2.0.0_systematic_noise.snv.bed.gz) is part of the standard DRAGEN Resource files available for DRAGEN v4.4 on Illumina's DRAGEN Secondary Analysis support site.
See under Illumina DRAGEN Somatic small variant calling - WGS, WES > SNV Somatic Systematic Noise v2.0.0
This is what is contained in the download .tar:

Is there demo data in BaseSpace at this moment or in near future?
In BaseSpace, demo data is available under the "Demo Data" tab.
In ICA, demo data is available in the DRAGEN v4.4 bundle under "/Illumina DRAGEN 5-Base Methylation Germline Demo Data".
For any feedback or questions regarding this article (Illumina Knowledge Article #9950), contact Illumina Technical Support [email protected].
Last updated
Was this helpful?
