Page History

...

Here is a helpful document from Illumina discussing the compatibility between various library primers and various sequencing platforms and kits.

...

1. Highlight
   color red
   P5 PCR primer/flowcell capture site:

   AATGATACGGCGACCACCGAGA

2. Highlight
   color yellow
   IndexRead2:

   NONE - as in do NOT put an index here.  If you want to add an index here, use one of the "Dual-index" designs below.

3. Highlight
   color green
   Read1 primer site:

   Either the small RNA sequencing primer site: (NEB: TCTACACGTTCAGAGTTCTACAGTCCGACGATCA [Illumina lists this but it is UNPROVEN: CAGGTTCAGAGTTCTACAGTCCGACGATCA]) OR the standard TruSeq Read 1 primer site: TCTACACTCTTTCCCTACACGACGCTCTTCCGATCT. Which to choose? The TruSeq Read 1 primer site is complementary to the Read 2 primer site, so if you are designing amplicons do NOT use the TruSeq Read 1 primer site, use the small RNA sequencing primer site.

4. The insert to be sequenced

5. Highlight
   color cyan
   Read2 primer site:

   Then the Index read primer site: AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC (NOTE: the initial A is from the dA tailing of the insert and is not included in the index primer or adaptor sequences; note also the reverse-complement of this is the Read 2 sequencing primer, but the Read 2 sequencing primer includes the T corresponding to the dA insert tail so sequencing starts with the insert)

6. Highlight
   color blue
   IndexRead1:

   The index sequence (usually 6 bp) - see many examples below in the Barcodes section. Within a lane, image analysis works best with as much base diversity as possible.

7. Highlight
   color purple
   P7 PCR primer/flowcell capture site:

   ATCTCGTATGCCGTCTTCTGCTTG

...

Here is an example of a read-pair from an RNA-seq library generated from the NEB small RNA kit with an insert size of 62 nt:

...

After exhaustive searching of all 4096 6-mers, the following table is all remaining 6 bp barcodes that have hamming distance of at least 3 from each other and the table above of 49 barcodes (NOTE: these have NOT been tested on the sequencer as of 2/7/12):

Excruciating details - USE WITH CAUTION - RNA PCR primers are NOT current as of Dec. 2011

...

Nextera adaptor style:

 5'-TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGT-3'
                  |||||||||||||||||||
3'-CAGAGCACCCGAGCCTCTACACATATTCTCTGTC-5'

TruSeq truncated adaptor:
  5'-ACACTCTTTCCCTACACGACGCTCTTCCGATCT-3'
                         |||||||||||||
 3'-CACTGACCTCAAGTCTGCACACGAGAAGGCTAGA-5'

Nextera adaptor style, with primers overlaid:
First cycle:
                                                                                                            3'-GGCTCGGGTGCTCTG CAAAGC TAGAGCATACGGCAGAAGACGAAC-5'
                                  5'-TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGT-3'   <UNKNOWN>   5'-CTGTCTCTTATACACATCTCCGAGCCCACGAGAC-3'
                                                   |||||||||||||||||||                      |||||||||||||||||||
                                 3'-CAGAGCACCCGAGCCTCTACACATATTCTCTGTC-5'    <UNKNOWN>  3'-TGACAGAGAATATGTGTAGACTGCGACGGCTGCT-5'


Second and subsequent cycles:
AATGATACGGCGACCACCGAGATCTACAC ATCACG TCGTCGGCAGCGTC
                                  3'-AGCAGCCGTCGCAGTCTACACATATTCTCTGTCA   <UNKNOWN>  3'-TGACAGAGAATATGTGTAGACTGCGACGGCTGCT-5'

...

Cautions, common mistakes, and lessons learned from failure

1. Assembling the P7 side adaptor or primer wrong - the key thing to note is that the "cannonical designs" are shown 5' to 3' across the entire finished sequencing construct.  So if you're designing a reverse primer for the P7 side you have to use the reverse complement of ALL 3 DESIGN ELEMENTS (flow cell binding site, barcode, and sequencing primer site) and make sure they're in the right order.
2. Incorrect P5 dual-index design - the "ACAC" motif in the single index design MUST be repeated on both sides of an index within P5 - see the "dual index" designs specifically.
3. Reverse complement barcode sequences in either P5 or P7 side indexes, especially from amplicons - the fact that the Illumina sequencers read i5 differently is a pain - pay attention to that when submitting barcode sequences that will wind up in a sample sheet.  And remember that the i7 index is read "forward, top strand" of the canonical design, which is reverse complement of the sequence that appears in a reverse primer used when creating a library.

...