I need help (or tips) for creating my own indexing-primers for Illumina seqeuncing (TrueSeq-compatible):
It seems to me that I could simply devise my own index primers based on the Trueseq adapter sequences and currently available index primer sequences, by simply exchanging the 8bp index with whatever I want. Then I could have these primers synthesized by e.g. Eurofins (or whatever), and use them instead of the indexing kits sold by Illumina and/or NEB, right?
Does anybody have experience in this? Are there any specific pitfalls to this, I should look out for? Do the Index-primers have to be modified in any way?
Background:
My lab is doing bacterial single cell and transcriptome sequencing, both of which produce uneven sequencing coverage.
Currently we are switching from sequencing on MiSeq & Nextseq-plattforms to using the HiSeq Xten, since we get much more output for the money (and have extremely good offers for that). With the number of libraries we need currently to sequence, using the older Illumina plattforms is definitively not feasible for us anymore.
However this Illumina plattform (and all subsequent new illumina plattforms as well) has a massive problem with "adapter hopping", due to the relatively new "ExAmp" cluster amplification method. This problem is even worse, to the point of being intolerable, when having uneven sequence coverage, such as transcriptomes or MDA'ed Single Cell sequencing libraries.
However, it seems this problem can be largely contained simply by using unique barcodes at each fragment end (currently in most dual indexed libraries there are only unique combinations, each reusing either the forward or the reverse index)
Sadly, the commercially available dual indexing kits for illumina libraries each offer only 8 forward and 12 reverse indices, allowing only for 8 double-unique indexed libraries (or up to 16 if you combine both of the available kits). This is not an option for us, as we have to multiplex at least 70 samples.
I do not understand this index limitation, as in theory it should be possible to create 4^8 = 65536 unique 8mer indices, so even when removing all extremely low-complexity options there should still be more than enough distinct 8mer-sequences left to create 192 unique indices (giving me 96 unique forward and 96 different unique reverse indices), right?
Even though this problem is receiving more and more attention these days, it seems no such option is commercially available yet. Therefore I need to create my own index-primers and want to make sure I take all necessary considerations into account.
It seems to me that I could simply devise my own index primers based on the Trueseq adapter sequences and currently available index primer sequences, by simply exchanging the 8bp index with whatever I want. Then I could have these primers synthesized by e.g. Eurofins (or whatever), and use them instead of the indexing kits sold by Illumina and/or NEB, right?
Does anybody have experience in this? Are there any specific pitfalls to this, I should look out for? Do the Index-primers have to be modified in any way?
Background:
My lab is doing bacterial single cell and transcriptome sequencing, both of which produce uneven sequencing coverage.
Currently we are switching from sequencing on MiSeq & Nextseq-plattforms to using the HiSeq Xten, since we get much more output for the money (and have extremely good offers for that). With the number of libraries we need currently to sequence, using the older Illumina plattforms is definitively not feasible for us anymore.
However this Illumina plattform (and all subsequent new illumina plattforms as well) has a massive problem with "adapter hopping", due to the relatively new "ExAmp" cluster amplification method. This problem is even worse, to the point of being intolerable, when having uneven sequence coverage, such as transcriptomes or MDA'ed Single Cell sequencing libraries.
However, it seems this problem can be largely contained simply by using unique barcodes at each fragment end (currently in most dual indexed libraries there are only unique combinations, each reusing either the forward or the reverse index)
Sadly, the commercially available dual indexing kits for illumina libraries each offer only 8 forward and 12 reverse indices, allowing only for 8 double-unique indexed libraries (or up to 16 if you combine both of the available kits). This is not an option for us, as we have to multiplex at least 70 samples.
I do not understand this index limitation, as in theory it should be possible to create 4^8 = 65536 unique 8mer indices, so even when removing all extremely low-complexity options there should still be more than enough distinct 8mer-sequences left to create 192 unique indices (giving me 96 unique forward and 96 different unique reverse indices), right?
Even though this problem is receiving more and more attention these days, it seems no such option is commercially available yet. Therefore I need to create my own index-primers and want to make sure I take all necessary considerations into account.
Comment