indexing on insert

[cli]

Dear all,

I would like to put double index on Illumina library to avoid PCR jumping and to pool samples together for gene capture. I remember there was a paper about putting index on the insert instead of on the adapters a few years back. I think it is better, because sequencing double index in not very popular, so it is hard to find enough samples for sharing a flow cell. I searched all my files and google sites and pulled all my hair and still can't find the paper I read before. Could anyone give me a clue of the paper I missed, and what do you think is the best strategy to solve the problem. Thanks in advance.

Chenhong

[Brian Bushnell]

Hmmm, you might search for "inline barcodes", which is what I have normally heard them called. What's "PCR jumping"?

I think it's kind of odd to ever use single-indexed barcodes, as dual-index reduces crosstalk, and I would have expected the added cost of dual indexing to be very small. Inline barcodes sound like a good solution to me. Just note that if they are not color-balanced, you'll reduce the quality of your cluster identification and genomic sequence; also, using inline barcodes means more manual work for demultiplexing (though you can still do that with BBDuk/Seal fairly automatically).

[nucacidhunter]

Dear all,

I would like to put double index on Illumina library to avoid PCR jumping and to pool samples together for gene capture. I remember there was a paper about putting index on the insert instead of on the adapters a few years back. I think it is better, because sequencing double index in not very popular, so it is hard to find enough samples for sharing a flow cell. I searched all my files and google sites and pulled all my hair and still can't find the paper I read before. Could anyone give me a clue of the paper I missed, and what do you think is the best strategy to solve the problem. Thanks in advance.

Chenhong

Addition of inline barcodes for applications beside PCR based and custom libraries is inconvenient. All the lanes in a flow cell can be set up to be sequenced as dual indxed and all Illumina sequencing kits has enough reagents for dual index sequencing. Second index reads for single indexed libraries can be masked during demultiplexing.

[atcghelix]

I agree that using inline barcodes is often inconvenient, and you do have to be careful about color balancing if you design them yourself. Also, keep in mind that you lose bases off of your R1/R2 if you use inline barcodes.

We usually prepare libraries by ligating universal y-forked stubs, then add the full-length adapters with dual indices via limited (~6) cycle PCR. Here's a paper that describes that: http://biorxiv.org/content/early/2016/06/15/049114

[cli]

Thanks for all answers.

I did found some information from a book chapter of my friends (Fortes & Johanna, 2015). An earlier paper talked about in-line indexing was Parkinson et al. (2012), but it used tagmentation protocol, not shearing. Anyway, here are the oligos I designed for indexed IS1/IS3 and IS2/IS3. I will order them today and report back here once we done the testing.

"pcr jumping" is cross talking. :-)

IS1_Ind1 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTtc*t*g*c*c
IS1_Ind2 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTgt*c*t*c*t
IS1_Ind3 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTat*a*t*t*g
IS1_Ind4 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTtg*g*a*a*g
IS1_Ind5 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTtc*t*a*g*t
IS1_Ind6 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTag*a*g*t*a
IS1_Ind7 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTgg*c*c*a*a
IS1_Ind8 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTta*t*c*t*c
IS1_Ind9 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTtt*a*t*g*c
IS1_Ind10 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTag*t*t*g*g
IS1_Ind11 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTgt*c*a*a*g
IS1_Ind12 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTca*g*c*a*a
IS1_Ind13 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTtc*g*c*c*g
IS1_Ind14 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTct*a*a*g*a
IS1_Ind15 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTcc*g*c*t*t
IS1_Ind16 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTaa*g*t*t*a
IS1_Ind17 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTgg*t*a*c*c
IS1_Ind18 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTcc*a*g*g*t
IS1_Ind19 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTaa*t*c*g*a
IS1_Ind20 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTaa*c*g*c*a
IS1_Ind21 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTga*c*g*a*c
IS1_Ind22 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTcg*c*g*c*t
IS1_Ind23 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTcc*g*t*a*g
IS1_Ind24 A*C*A*C*TCTTTCCCTACACGACGCTCTTCCGATCTgt*a*a*t*c
IS2_Ind25 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTga*c*c*t*t
IS2_Ind26 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTtc*a*t*a*a
IS2_Ind27 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTca*a*g*a*g
IS2_Ind28 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTcg*a*t*c*a
IS2_Ind29 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTtt*g*a*t*t
IS2_Ind30 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTtc*c*g*a*g
IS2_Ind31 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTcc*t*g*a*a
IS2_Ind32 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTat*t*c*t*t
IS2_Ind33 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTgc*g*a*c*t
IS2_Ind34 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTgg*c*t*t*c
IS2_Ind35 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTaa*t*a*c*g
IS2_Ind36 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTta*c*g*g*t
IS2_Ind37 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTac*c*g*t*c
IS2_Ind38 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTag*a*a*g*c
IS2_Ind39 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTca*t*a*g*c
IS2_Ind40 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTag*g*c*t*c
IS2_Ind41 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTct*g*c*g*g
IS2_Ind42 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTct*c*g*g*c
IS2_Ind43 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTga*t*t*a*g
IS2_Ind44 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTag*a*t*a*t
IS2_Ind45 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTtg*g*t*c*c
IS2_Ind46 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTgt*t*c*c*g
IS2_Ind47 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTgt*a*c*g*t
IS2_Ind48 G*T*G*A*CTGGAGTTCAGACGTGTGCTCTTCCGATCTaa*g*a*a*c
IS3_Ind1 ggcagaAGATCGGAA*G*A*G*C
IS3_Ind2 agagacAGATCGGAA*G*A*G*C
IS3_Ind3 caatatAGATCGGAA*G*A*G*C
IS3_Ind4 cttccaAGATCGGAA*G*A*G*C
IS3_Ind5 actagaAGATCGGAA*G*A*G*C
IS3_Ind6 tactctAGATCGGAA*G*A*G*C
IS3_Ind7 ttggccAGATCGGAA*G*A*G*C
IS3_Ind8 gagataAGATCGGAA*G*A*G*C
IS3_Ind9 gcataaAGATCGGAA*G*A*G*C
IS3_Ind10 ccaactAGATCGGAA*G*A*G*C
IS3_Ind11 cttgacAGATCGGAA*G*A*G*C
IS3_Ind12 ttgctgAGATCGGAA*G*A*G*C
IS3_Ind13 cggcgaAGATCGGAA*G*A*G*C
IS3_Ind14 tcttagAGATCGGAA*G*A*G*C
IS3_Ind15 aagcggAGATCGGAA*G*A*G*C
IS3_Ind16 taacttAGATCGGAA*G*A*G*C
IS3_Ind17 ggtaccAGATCGGAA*G*A*G*C
IS3_Ind18 acctggAGATCGGAA*G*A*G*C
IS3_Ind19 tcgattAGATCGGAA*G*A*G*C
IS3_Ind20 tgcgttAGATCGGAA*G*A*G*C
IS3_Ind21 gtcgtcAGATCGGAA*G*A*G*C
IS3_Ind22 agcgcgAGATCGGAA*G*A*G*C
IS3_Ind23 ctacggAGATCGGAA*G*A*G*C
IS3_Ind24 gattacAGATCGGAA*G*A*G*C
IS3_Ind25 aaggtcAGATCGGAA*G*A*G*C
IS3_Ind26 ttatgaAGATCGGAA*G*A*G*C
IS3_Ind27 ctcttgAGATCGGAA*G*A*G*C
IS3_Ind28 tgatcgAGATCGGAA*G*A*G*C
IS3_Ind29 aatcaaAGATCGGAA*G*A*G*C
IS3_Ind30 ctcggaAGATCGGAA*G*A*G*C
IS3_Ind31 ttcaggAGATCGGAA*G*A*G*C
IS3_Ind32 aagaatAGATCGGAA*G*A*G*C
IS3_Ind33 agtcgcAGATCGGAA*G*A*G*C
IS3_Ind34 gaagccAGATCGGAA*G*A*G*C
IS3_Ind35 cgtattAGATCGGAA*G*A*G*C
IS3_Ind36 accgtaAGATCGGAA*G*A*G*C
IS3_Ind37 gacggtAGATCGGAA*G*A*G*C
IS3_Ind38 gcttctAGATCGGAA*G*A*G*C
IS3_Ind39 gctatgAGATCGGAA*G*A*G*C
IS3_Ind40 gagcctAGATCGGAA*G*A*G*C
IS3_Ind41 ccgcagAGATCGGAA*G*A*G*C
IS3_Ind42 gccgagAGATCGGAA*G*A*G*C
IS3_Ind43 ctaatcAGATCGGAA*G*A*G*C
IS3_Ind44 atatctAGATCGGAA*G*A*G*C
IS3_Ind45 ggaccaAGATCGGAA*G*A*G*C
IS3_Ind46 cggaacAGATCGGAA*G*A*G*C
IS3_Ind47 acgtacAGATCGGAA*G*A*G*C
IS3_Ind48 gttcttAGATCGGAA*G*A*G*C

[Brian Bushnell]

This is a little confusing to me; I've never designed or ordered oligos. Can you explain what "*" means, why some letters are lower-case, and the meaning of "IS1"/"IS2"/"IS3"? It seems like some of the bases are for binding, and some are indexes, but I'm not really sure. I'm guessing IS1 and IS2 are in the normal index cycles and IS3 is the inline one... but in that case I don't understand why there is an identical "AGATCGGAA" in each IS3.

[cli]

* means PTO bond to stabilize the end of the oligos. The low-case letters indicate the indices themselves. IS1 and IS3 are reverse complemental to each other, so they will be annealed and form a double strand adapter. IS2 and IS3 will from the other adapter. Then we can add the pair of adapters with different index on to different samples and the ligation step of library prep. At least that's my plan.