Index Orientation Guide

Guidelines for reverse complementing i5 sequences for demultiplexing

When performing dual indexing, all Illumina sequencers read the first index (i7) in the same direction; however, the second index (i5) is read in different orientations depending on the instrument and chemistry, in particular, differences between when the paired-end turnaround and i5 read take place:

• On the MiSeq, HiSeq 2000/2500, and the MiniSeq Rapid kits the i5 sequence is read in the forward orientation.

• On the iSeq 100, MiniSeq, NextSeq 500/500, NextSeq 1000/2000, HiSeq 3000/4000/X systems, NovaSeq 6000 (v1.5 reagents), and the NovaSeq X/X Plus the i5 sequence is read in the reverse complement orientation.

See the Index Orientation Table for information on when to use which orientation of the i5 sequence for these platforms depending on the analysis software used.

When demultiplexing, the final sample sheet provided to the FASTQ generation software needs to have the i5 index in the correct orientation for that platform. However, most Illumina sample sheet generation software handles this automatically. This means that these software work on the assumption that your i5 sequence is always in the forward orientation and will automatically create the reverse complement of the sequences provided for the appropriate instrument type.

The following provides guidance on when to reverse complement and when to not reverse complement the i5 sequences when analyzing iSeq, MiniSeq, NextSeq 500/550, NextSeq 1000/2000, HiSeq 3000/4000/X, NovaSeq 6000 (v1.5 reagents), or NovaSeq X/X Plus data.

Run Planning

BaseSpace Run Planning

For Run Planning in BaseSpace, the output is always a v2 sample sheet for BCL Convert:

• NextSeq 1000/2000 and NovaSeq X/X Plus i5 indexes will correctly be output in the forward orientation for later downstream processing. Custom kits should also have the i5 indexes entered in the forward orientation.

• For other supported platforms (such as the NovaSeq 6000), Run Planning will add a line to the Header section of the v2 sample sheet (IndexOrientation,Forward) and the i5 will be written in the forward orientation and processed correctly downstream.

• If generating v2 sample sheets for instruments other than the ones listed, confirm the final sample sheet has the i5 index in the correct orientation as described above.

Manual Run Upload

Runs Uploaded Manually

For runs uploaded manually (i.e., an Illumina tool is not being used to generate the sample sheet), the index orientation depends on which instrument you are using. Please see the Index Orientation Table for information on how to enter the i5 sequence depending on the instrument and software used.

IEM

IEM

When using IEM, i5 sequences must be entered into these software in the forward orientation in all circumstances and for all instrument types.

The sequences displayed are what is entered into the final sample sheet. Kit definitions have the i5 in the forward orientation. The i5 sequences will be appropriately reverse complemented and displayed in the IEM interface based on instrument/reagent type. IEM will use the instrument/reagent type selected in the setup wizard to generate the final sample sheet with the i5 sequences in the correct orientation.

PrepTab

BaseSpace PrepTab

When using BaseSpace Sequence Hub PrepTab, i5 sequences must be entered into these software in the forward orientation in all circumstances and for all instrument types.

The software automatically creates the reverse complement as needed for analysis. If the sequence is reverse complemented, this will only be shown in the SampleSheetUsed.csv file.

LRM

LRM

When using LRM, i5 sequences must be entered into these software in the forward orientation in all circumstances and for all instrument types.

The software automatically creates the reverse complement as needed for analysis. If the sequence is reverse complemented, this will only be shown in the SampleSheetUsed.csv file or the SampleSheet.csv file created by LRM at the analysis start. LRM will detect the instrument type from the run files and create a final sample sheet with the i5 indexes in the correct orientation for that platform.

Important Caveats for MiniSeq Rapid Kits

• The MiniSeq Rapid made a recipe change that uses the grafted oligo for the i5 index priming, so it is read in the forward orientation similar to a MiSeq.

• If running in manual mode and using bcl2fastq or BCL Convert, or uploading the sample sheet to BaseSpace as a manual mode run and attaching a sample sheet, enter the i5 sequence in the forward direction. Runs uploaded in manual mode have their data directly passed to bcl2fastq with no further processing.

• For Local Run Manager (LRM), input the sequences in the reverse complement when setting up the run so the software automatically creates the forward orientation as needed for analysis. This is the opposite of most of our guidance and is only for the MiniSeq Rapid kits.

Color Balance Considerations

When assessing index sets for color balance, the sequence observed by the instrument must be assessed, regardless of the orientation used for the sample sheet. For the i7 index, always asses color balance using the "i7 bases for sample sheet" column orientation. For the i5 index, assess color balance using the forward orientation for MiSeq runs and MiniSeq runs with rapid reagents, and the reverse-complement orientation for all other instruments.

Which orientation are the indexes in Top_Unknown_Barcodes output from BclConvert?

For The NextSeq 1000/2000 and NovaSeq X/X Plus platforms, the RunInfo.xml file includes the IsReverseComplement flag. With this flag, the index sequences that BCL Convert recognizes and outputs are in the following orientations in the Top_Unknown_Barcodes output file.

• Index 1: the i7 index column is in the forward orientation (match the orientation if the i7 indexes in the sample sheet).

• Index 2: the i5 index column is in the reverse complement orientation (are the reverse complement of the i5 indexes in the sample sheet).

For all instruments, the indexes shown in the Top_Unknown_Barcodes are the sequences as they were sequenced/read by the sequencer.