Hi all, this may be a huge FAQ... Does anybody know scientific evidences to choose a HiSeq2000 over Solid 5500xl instrument? Which real advantage I can get by sequencing 100 bp in paired ends? Don't want to start a flame.
d
I'm not sure what scientific evidence you're looking for, there doesn't seem to be many papers comparing the technologies, and which technology is best probably depends a lot on the application it is intended for.
For instance, mate pair construction is tricky, and I suspect that SOLiD, which used a separate primer for each mate will let you identify more correct mates, compared to Illumina, where you get a pair even if the mate construction failed (i.e. the fragment doesn't contain the adapter).
My experience is limited to Illumina for now, and it's my impression that HiSeq2k improves quite a bit on the previous platforms.
I also have experience with Illumina, but now I have to justify why I would need the 100bp-PE capability of the HiSeq2k. I mean, in terms of precision, Illumina relies on the high number and the longer (paired) reads (up to 100/150 bp in paired ends), Solid relies on its chemistry... in the end which is better?
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
It is possible that Q&A site for SeqAnswers forum might be a better venue for hardware questions - have a look at http://i.seqanswers.com/
I believe dawe is in that forum, but i.seqanswers.com currently is not as responsive as this forum. The seqanswers.com is really active, though.
It depends on what you intend to do. In general, I am inclined to hiseq. It is not only about read length. Hiseq data are also cleaner and easier/more efficient to process, in comparison to SOLiD3 (I do not know about 5500xl).
I would suggest adding another dimension to your question. The sequence generated by these machines needs to be analyzed. For every group that I know of, the downstream processing of the data is the real issue and, for better or worse, there is more choice of algorithms and packages for Illumina data. I do not think this point should be an afterthought....