DRAGEN Array Methylation QC

Methylation QC Threshold Adjustment

When using DRAGEN Array – Methylation – QC cloud analysis type, additional customization options will appear after product files are selected within Configuration Settings. Adjustments to these thresholds will be saved as part of the Configuration Setting. Thresholds can be adjusted based on study objectives. Adjusting thresholds will impact the pass or fail status of samples in the output files.

Illumina recommends thresholds for MethylationEPIC v1 & v2 and Methylation Screening Array (MSA). Users may use these thresholds as a starting point when defining thresholds for their custom or semi-custom BeadChip or other Infinium Methylation arrays. Further tuning may be required based on BeadChip used, laboratory conditions, iScan settings, bisulfite conversion methods, FPPE sample type, etc. A dataset deemed acceptable to the user based on proportion probes passing can be used for these additional threshold adjustments.

To customize thresholds, use the toggle to allow additional thresholds to be displayed and adjust as desired by typing in a numeric value or using the arrows to adjust up or down. Further detail of these thresholds including calculation method can be found in the Methylation Sample QC Summary Files section.

The recommended thresholds are pre-set within the software for MethylationEPIC and Methylation Screening Array with the following values:

Threshold	Methylation Screening Array	MethylationEPIC
	0	0
StainingGreen	5	5
StainingRed	5	5
ExtensionGreen	5	5
ExtensionRed	5	5
HybridizationHighMedium	1	1
HybridizationMediumLow	1	1
TargetRemoval1	1	1
TargetRemoval2	1	1
BisulfiteConversion1Green	1	1
BisulfiteConversion1BackgroundGreen	0.5	1
BisulfiteConversion1Red	1	1
BisulfiteConversion1BackgroundRed	0.5	1
BisulfiteConversion2	0.5	1
BisulfiteConversion2Background	0.5	1
Specificity1Green	1	1
Specificity1Red	1	1
Specificity2	1	1
Specificity2Background	1	1
NonpolymorphicGreen	2.5	5
NonpolymorphicRed	3	5
BgCorrectionOffset	3000	3000
PvalThreshold	0.05	0.05

The first 21 rows in the tables correspond to the 21 control metrics used in the methylation sample QC. See section Methylation Sample QC Summary Files for details.

DRAGEN Array Methylation QC and GenomeStudio Methylation Module Differences

DRAGEN Array Methylation QC software provides automated methylation sample QC using assay control probes on the Infinium Methylation Arrays. Unlike the manual visual QC in GenomeStudio, DRAGEN Array ultilizes 21 numerical metrics defined based on the control probes and uses standard thresholds to determine pass/fail status of a sample. Unlike GenomeStuio, probe detection rate (proportion of probes passing at a given p-value threshold) is not utilized to determine sample pass/fail status in DRAGEN Array. For more information, see High-throughput Infinium methylation array QC using DRAGEN Array Methylation QC software tech note.

DRAGEN Array Methylation QC performs background normalization, dye bias correction, and detection p-value calculation differently in comparison to the GenomeStudio Methylation module, leading to differences in probe detection p-values and detection rates. For the GenomeStudio Methylation Module, non-cancer samples at standard DNA input typically have detection rate > 96%. The detection rates from DRAGEN Array Methylation QC are typically lower compared to GenomeStudio, because the detection p-value from DRAGEN Array is more stringent than that from the GenomeStudio Methylation Module. The table below shows example detection rates from the DRAGEN Array Methylation QC software from MSA (Methylation Screening Array) datasets.

Dataset	Min detection rate	Mean detection rate	Sample Count
A	86%	93%	220
B	61%	83%	951
C	63%	85%	34
D	77%	85%	22

Note that only samples passing QC are included and all samples are at or above 50ng DNA input. Detection p-value threshold 0.05.