s3-ATAC

Due to semi-suppressive PCR (click here to see a detailed explanation), only 50% of the tagmented material from the regular Tn5 are amplifiable. The s3-ATAC approach introduced uracil bases in the Tn5 adapters and used uracil intolerant and tolerant DNA polymerases for the gap fill-in and PCR amplification, respectively. In this way, every Tn5 cutting event is amplifiable. Combined with the combinatorial indexing strategy, it is a very clever technique that significantly increases the library complexities of many single cell libraries with a relatively easy workflow.

Adapter and primer sequences:

Step-by-step library generation

(1) Anneal Barcoded Truseq_R2_SBS12_partial and Tn5 binding site 19-bp Mosaic End (ME) bottom strand to assemble Tn5 transposome (Tn5 homodimer):

(2) Sort certain number of nuclei into wells, and perform tagmentation using barcoded Tn5 transposome:

Only one product, the same sequence at both ends:

5'- CGTGTGCTCTTCCGATCT[8-bp Tn5 index]/ideoxyU/AGATGTGTATAAGAGACAGXXXXXXXXXXXX...XXX         CTGTCTCTTATACACATCT
                                               TCTACACATATTCTCTGTC         XXX...XXXXXXXXXXXXGACAGAGAATATGTGTAGA/ideoxyU/[8-bp Tn5 index]TCTAGCCTTCTCGTGTGC -5'

(3) Pool all wells, re-distribute into wells in a new plate and perform gap fill-in using a uracil-intolerant DNA polymerase (e.g. the NPM from the Nextera XT kit):

5'- CGTGTGCTCTTCCGATCT[8-bp Tn5 index]/ideoxyU/AGATGTGTATAAGAGACAGXXXXXXXXXXXX...XXXXXXXXXXXXCTGTCTCTTATACACATCT
                                               TCTACACATATTCTCTGTCXXXXXXXXXXXX...XXXXXXXXXXXXGACAGAGAATATGTGTAGA/ideoxyU/[8-bp Tn5 index]TCTAGCCTTCTCGTGTGC -5'

(4) Add the A14_ME_LNA oligo for adapter switching:

The top strand:

5'- CGTGTGCTCTTCCGATCT[8-bp Tn5 index]/ideoxyU/AGATGTGTATAAGAGACAGXXXXXXXXXXXX...XXXXXXXXXXXXCTGTCTCTTATACACATCT------------->
                                                                                            /GACAGAGAATATGTGTAGACTGCGACGGCTGCT -5'
                                                                                      3InvdT

The bottom strand:
                                     /3InvdT/
5'- TCGTCGGCAGCGTCAGATGTGTATAAGAGACAG
    <-------------TCTACACATATTCTCTGTCXXXXXXXXXXXX...XXXXXXXXXXXXGACAGAGAATATGTGTAGA/ideoxyU/[8-bp Tn5 index]TCTAGCCTTCTCGTGTGC -5'

(5) This is the amplifiable products after adapter switching (Note that the products from the top and bottom strands are the same):

3'- AGCAGCCGTCGCAGTCTACACATATTCTCTGTCXXXXXXXXXXXX...XXXXXXXXXXXXGACAGAGAATATGTGTAGA/ideoxyU/[8-bp Tn5 index]TCTAGCCTTCTCGTGTGC -5'

(6) Add PCR_i5_primer and PCR_i7_primer and use a uracil-tolerant polymerase (e.g. NEBNext Q5U) for the library amplification:

5'- AATGATACGGCGACCACCGAGATCTACAC[i5]TCGTCGGCAGCGTC------------->
                                 3'- AGCAGCCGTCGCAGTCTACACATATTCTCTGTCXXXXXXXXXXXX...XXXXXXXXXXXXGACAGAGAATATGTGTAGA/ideoxyU/[8-bp Tn5 index]TCTAGCCTTCTCGTGTGC -5'
                                                                                                                              <--------------TCTAGCCTTCTCGTGTGCAGACTTGAGGTCAGTG[i7]TAGAGCATACGGCAGAAGACGAAC -5'

(7) Final library structure (cell barcodes are the combination of the 8-bp Tn5 + 8-bp i7 + 8-bp i5 barcodes):

5'- AATGATACGGCGACCACCGAGATCTACACNNNNNNNNTCGTCGGCAGCGTCAGATGTGTATAAGAGACAGXXXXXX...XXXXXXCTGTCTCTTATACACATCTANNNNNNNNAGATCGGAAGAGCACACGTCTGAACTCCAGTCACNNNNNNNNATCTCGTATGCCGTCTTCTGCTTG -3'
3'- TTACTATGCCGCTGGTGGCTCTAGATGTGNNNNNNNNAGCAGCCGTCGCAGTCTACACATATTCTCTGTCXXXXXX...XXXXXXGACAGAGAATATGTGTAGATNNNNNNNNTCTAGCCTTCTCGTGTGCAGACTTGAGGTCAGTGNNNNNNNNTAGAGCATACGGCAGAAGACGAAC -5'
            Illumina P5            i5           s5              ME             gDNA               ME           8 bp             Truseq Read 2              i7          Illumina P7
                                                                                                            Tn5 barcode                                                                             

Library sequencing:

(1) Add read 1 sequencing primer to sequence the first read (bottom strand as template, these are the gDNA reads, 85 cycles):

                                     5'- TCGTCGGCAGCGTCAGATGTGTATAAGAGACAG------------->
3'- TTACTATGCCGCTGGTGGCTCTAGATGTGNNNNNNNNAGCAGCCGTCGCAGTCTACACATATTCTCTGTCXXXXXX...XXXXXXGACAGAGAATATGTGTAGATNNNNNNNNTCTAGCCTTCTCGTGTGCAGACTTGAGGTCAGTGNNNNNNNNTAGAGCATACGGCAGAAGACGAAC -5'

(2) Add Index 1 sequencing primer to sequence the i7 index (bottom strand as template, 10 cycles were used in the paper, the first 8 bp are i7):

                                                                                                                  5'- GATCGGAAGAGCACACGTCTGAACTCCAGTCAC--------->
3'- TTACTATGCCGCTGGTGGCTCTAGATGTGNNNNNNNNAGCAGCCGTCGCAGTCTACACATATTCTCTGTCXXXXXX...XXXXXXGACAGAGAATATGTGTAGATNNNNNNNNTCTAGCCTTCTCGTGTGCAGACTTGAGGTCAGTGNNNNNNNNTAGAGCATACGGCAGAAGACGAAC -5

(3) Cluster regeneration, add Index 2 sequencing primer to sequence the i5 index (top strand as template, 10 cycles were used in the paper, the first 8 bp are i5):

5'- AATGATACGGCGACCACCGAGATCTACACNNNNNNNNTCGTCGGCAGCGTCAGATGTGTATAAGAGACAGXXXXXX...XXXXXXCTGTCTCTTATACACATCTANNNNNNNNAGATCGGAAGAGCACACGTCTGAACTCCAGTCACNNNNNNNNATCTCGTATGCCGTCTTCTGCTTG -3'
                               <---------AGCAGCCGTCGCAGTCTACACATATTCTCTGTC -5'

(4) Add Read 2 sequencing primer to sequence the second read (top strand as template, 85 cycles, the first 8 cycles are the Tn5 barcodes):

5'- AATGATACGGCGACCACCGAGATCTACACNNNNNNNNTCGTCGGCAGCGTCAGATGTGTATAAGAGACAGXXXXXX...XXXXXXCTGTCTCTTATACACATCTANNNNNNNNAGATCGGAAGAGCACACGTCTGAACTCCAGTCACNNNNNNNNATCTCGTATGCCGTCTTCTGCTTG -3'
                                                                                    <--------------------------------TCTAGCCTTCTCGTGTGCAGACTTGAGGTCAGTG -5'