The length of each dna sequence including the primer would be 70bp long. I will have 1000 of them in many copies, where I have to detect all the 1000 in a single sequencing reaction (one sample with 1000 dna seqs). Is this possible with the current gen sequencing. please let me know>

Hi Sandeep,

The idea is very similar to the barseq: sequencing of thousand of 6 to 12 mers barcodes

http://genome.cshlp.org/content/early/2009/07/21/gr.093955.109

combined with Miseq run you can sequence up to 150b in one read, and got 30 millions reads per run.

Celia

The lenght of 70 bp might cause problems on a NGS platform, like the MiSeq or Ion Torrent PGM, a microRNA protocol might work in this case. A Solid platform might be a better choise, this platform has reads of 75 bp.

Martin you are right, we found that the min size of our library should be at least 250-350 on a Miseq, but it seems that our 150nt library works fine on a GAII.

Sandeep can you redesign one of the primer to increase your library size?

First of all I thank you all for giving me a viable solution. Celia, I can increase the size of each DNA barcode however, this will increase the overall cost of my project. To make it clear, the barcode is only 20bps long but the pimer is 25 bp (50 F+R) long so, is there any way (MiSeq or PGM) I could go with the 20bp barcode design such that I will have a successful result. If the above is impossible then what would be my best bet in terms of designing the barcode?

The barcode for the barseq data I gave you the paper, is only 20bp too, what we did to sequence our library on the Miseq is to move one of the oligo more downstream so our PCR product size will be 250bp. The problem with the miseq is not the read length, you can ask for a 20bp run if you like. The problem with shorter library is the cluster formation, you need to have a proper cluster formation step a longer library.

The price for seq will be the same if you only sequence from one side.

Celia thank you for your kind advice. It is not about the cost of sequencing but for the synthesis of 40,000 different barcodes, the group has to shed a lot of money. However, Was each barcode used in MiSeq paper integrated in the host genome or it just lies inside the cell. With my case each specific barcode just lies in the cytosol and I pool specific phenotype displaying cells and sequence them for the number of copies of each barcode; implying the players responsible for the phenotype (in a nut shell). Do you think I still need to increase the size of the primer or barcode just to get a robust read out of sequencing. Please let me know. Thank you.

If you have to synthesize a lot of different barcodes, is it an idea, like some researchers already have proven before to work, to add a random barcode sequence. This means you syntesize one oligo of 250 bp, but at the beginning or end you add 10 or 20 N's. These N's will generate random barcodes and you have to order just one oligo. I have seen poeple using a sequence like this : AAGCAGtGGTAtCAACGCAGAGtNNNNtNNNNtNNNNtCTT

Thank you Martin for your kind reply. I have written a computer algorithm to generate random barcodes of any length and flank them with primers. However I am not sure how long does each barcode (including common primer) needed to be such that I can sequence them easily via the NGS (MiSeq, PGM). The lower the better for saving costs of producing them, however I will count on you for your opinion on this query. Cheers.