txci-ATAC-seq

In txci-ATAC-seq, nuclei were first tagged with different indexed Tn5 in bulk and then were overloaded into the 10X Genomics Single Cell ATAC platform. In this way, you get way more nuclei compared to a regular 10x scATAC-seq run. The idea is like dsciATAC-seq, but txci-ATAC-seq works on the 10X Genomics platform, and custom sequencing primers are not needed. The procedures described here are based on their Protocols.io page.


Adapter and primer sequences:

10x Genomics Beads-oligo: |--5'- AATGATACGGCGACCACCGAGATCTACAC[16-bp GEM barcode]TCGTCGGCAGCGTC -3'

Tn5ME-A: 5'- TCGTCGGCAGCGTCAGATGTGTATAAGAGACAG -3'

Tn5ME-B: 5'- CGTGTGCTCTTCCGATCT[8-bp Tn5 barcode]AGATGTGTATAAGAGACAG -3'

* The barcode sequence on the 96w plate layout can be found from this file.

Tn5MErev: 5'-/phos/ CTGTCTCTTATACACATCT -3'

Nextera Tn5 binding site (19-bp Mosaic End (ME)): 5'- AGATGTGTATAAGAGACAG -3'

Nextera S5xx primer entry point (s5): 5'- TCGTCGGCAGCGTC -3'

Short SBS primer (TruSeq Read 2 primer entry point): 5'- CGTGTGCTCTTCCGATCT -3'

Custom i7 TruSeq primer: 5'- CAAGCAGAAGACGGCATACGAGAT[8-bp i7]GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCT -3'

Illumina Nextera Read 1 primer: 5'- TCGTCGGCAGCGTCAGATGTGTATAAGAGACAG -3'

Illumina TruSeq Read 2 sequencing primer: 5'- GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCT -3'

Sample Index sequencing primer: 5'- GATCGGAAGAGCACACGTCTGAACTCCAGTCAC -3'

GEM barcode sequencing primer: 5'- CTGTCTCTTATACACATCTGACGCTGCCGACGA -3'

Illumina P5 adapter: 5'- AATGATACGGCGACCACCGAGATCTACAC -3'

Illumina P7 adapter: 5'- CAAGCAGAAGACGGCATACGAGAT -3'

Step-by-step library generation:

(1) Anneal "Tn5MErev+Tn5ME-A" and "Tn5MErev+Tn5ME-B" and use those to assemble the indexed Tn5 transposomes:

Tn5 dimer

(2) Bulk nuclei tagging by the indexed Tn5 shown above. There are 3 different products (will create 9 bp gap):


Product 1 (s5 at both ends, not amplifiable due to semi-suppressiev PCR):

5'- TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGXXXXXXXXXXXX...XXX         CTGTCTCTTATACACATCT
                  TCTACACATATTCTCTGTC         XXX...XXXXXXXXXXXXGACAGAGAATATGTGTAGACTGCGACGGCTGCT -5'


Product 2 (s7 at both ends, not amplifiable due to semi-suppressiev PCR):

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


Product 3 (different ends, amplifiable):

5'- TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGXXXXXXXXXXXX...XXX         CTGTCTCTTATACACATCT
                  TCTACACATATTCTCTGTC         XXX...XXXXXXXXXXXXGACAGAGAATATGTGTAGA[8-bp Tn5 barcode]TCTAGCCTTCTCGTGTGC -5'

(3) Load to 10X machine, add SBS Short Primer to directly perform exponential PCR (not linear) which was shown to be the most effective to avoid barcode swapping:


Gap fill-in (72 C, 5 mins):

5'- TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGXXX...XXXCTGTCTCTTATACACATCT[8-bp Tn5 barcode]AGATCGGAAGAGCACACG -3'
3'- AGCAGCCGTCGCAGTCTACACATATTCTCTGTCXXX...XXXGACAGAGAATATGTGTAGA[8-bp Tn5 barcode]TCTAGCCTTCTCGTGTGC -5'


Exponential PCR in GEM:

|--5'- AATGATACGGCGACCACCGAGATCTACAC[16-bp GEM barcode]TCGTCGGCAGCGTC---------->
                                                   5'- TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGXXX...XXXCTGTCTCTTATACACATCT[8-bp Tn5 barcode]AGATCGGAAGAGCACACG -3'
                                                   3'- AGCAGCCGTCGCAGTCTACACATATTCTCTGTCXXX...XXXGACAGAGAATATGTGTAGA[8-bp Tn5 barcode]TCTAGCCTTCTCGTGTGC -5'
                                                                                                                           <----------TCTAGCCTTCTCGTGTGC -5'

(4) Break emulsion and DNA purification. This is the product from above:


5'- AATGATACGGCGACCACCGAGATCTACAC[16-bp GEM barcode]TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGXXX...XXXCTGTCTCTTATACACATCT[8-bp Tn5 barcode]AGATCGGAAGAGCACACG
    TTACTATGCCGCTGGTGGCTCTAGATGTG[16-bp GEM barcode]AGCAGCCGTCGCAGTCTACACATATTCTCTGTCXXX...XXXGACAGAGAATATGTGTAGA[8-bp Tn5 barcode]TCTAGCCTTCTCGTGTGC -5'

(5) Sample Index PCR using Illumina P5 and Custom i7 TruSeq primer:


5'- AATGATACGGCGACCACCGAGATCTACAC------------------->
5'- AATGATACGGCGACCACCGAGATCTACAC[16-bp GEM barcode]TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGXXX...XXXCTGTCTCTTATACACATCT[8-bp Tn5 barcode]AGATCGGAAGAGCACACG
    TTACTATGCCGCTGGTGGCTCTAGATGTG[16-bp GEM barcode]AGCAGCCGTCGCAGTCTACACATATTCTCTGTCXXX...XXXGACAGAGAATATGTGTAGA[8-bp Tn5 barcode]TCTAGCCTTCTCGTGTGC -5'
                                                                                                               <-------------------TCTAGCCTTCTCGTGTGCAGACTTGAGGTCAGTG[8-bp i7]TAGAGCATACGGCAGAAGACGAAC -5'

(6) Final library structure:


5'- AATGATACGGCGACCACCGAGATCTACACNNNNNNNNNNNNNNNNTCGTCGGCAGCGTCAGATGTGTATAAGAGACAGXXX...XXXCTGTCTCTTATACACATCTNNNNNNNNAGATCGGAAGAGCACACGTCTGAACTCCAGTCACNNNNNNNNATCTCGTATGCCGTCTTCTGCTTG
    TTACTATGCCGCTGGTGGCTCTAGATGTGNNNNNNNNNNNNNNNNAGCAGCCGTCGCAGTCTACACATATTCTCTGTCXXX...XXXGACAGAGAATATGTGTAGANNNNNNNNTCTAGCCTTCTCGTGTGCAGACTTGAGGTCAGTGNNNNNNNNTAGAGCATACGGCAGAAGACGAAC -5'
             Illumina P5              16 bp           s5              ME            gDNA          ME            8 bp            TruSeq Read 2             8 bp        Illumina P7
                                   GEM barcode                                                              Tn5 barcode                               sample index

Library sequencing:

(1) Add Nextera Read 1 sequencing primer to sequence the first read (51 cycles, bottom strand as template, gDNA read):


                                             5'- TCGTCGGCAGCGTCAGATGTGTATAAGAGACAG------->
3'- TTACTATGCCGCTGGTGGCTCTAGATGTGNNNNNNNNNNNNNNNNAGCAGCCGTCGCAGTCTACACATATTCTCTGTCXXX...XXXGACAGAGAATATGTGTAGANNNNNNNNTCTAGCCTTCTCGTGTGCAGACTTGAGGTCAGTGNNNNNNNNTAGAGCATACGGCAGAAGACGAAC -5'

(2) Add Sample Index sequencing primer to sequence sample index (i7) (bottom strand as template, 8 cycles):


                                                                                                                  5'- GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCT------->
3'- TTACTATGCCGCTGGTGGCTCTAGATGTGNNNNNNNNNNNNNNNNAGCAGCCGTCGCAGTCTACACATATTCTCTGTCXXX...XXXGACAGAGAATATGTGTAGANNNNNNNNTCTAGCCTTCTCGTGTGCAGACTTGAGGTCAGTGNNNNNNNNTAGAGCATACGGCAGAAGACGAAC -5'

(3) Cluster regeneration, add GEM barcode sequencing primer to sequence the second index (i5) (top strand as template, 16 cycles, this is the GEM barcode):


5'- AATGATACGGCGACCACCGAGATCTACACNNNNNNNNNNNNNNNNTCGTCGGCAGCGTCAGATGTGTATAAGAGACAGXXX...XXXCTGTCTCTTATACACATCTNNNNNNNNAGATCGGAAGAGCACACGTCTGAACTCCAGTCACNNNNNNNNATCTCGTATGCCGTCTTCTGCTTG
                                 <---------------AGCAGCCGTCGCAGTCTACACATATTCTCTGTC -5'

(4) Add TruSeq Read 2 primer to sequence the second read (top strand as template, 78 cycles, the first 8 bp are the Tn5 barcodes, the last 51 bp are gDNA, the middle 19 bp are ME):


5'- AATGATACGGCGACCACCGAGATCTACACNNNNNNNNNNNNNNNNTCGTCGGCAGCGTCAGATGTGTATAAGAGACAGXXX...XXXCTGTCTCTTATACACATCTNNNNNNNNAGATCGGAAGAGCACACGTCTGAACTCCAGTCACNNNNNNNNATCTCGTATGCCGTCTTCTGCTTG
                                                                                      <-------------------------------TCTAGCCTTCTCGTGTGCAGACTTGAGGTCAGTG -5'