MiSeq maximum fragment size

[Peter Meintjes]

Hi there, does anyone know if there is a maximum size fragment that won't be sequenced by a MiSeq? The approach is paired end 250x250 on a size selected library of fragments. I'd like to use size selection to sequence a 500bp range of fragments (i.e. 900-1400 or 1200-1700 or 1500-2000), but have heard that fragments over ~1200 don't get sequenced. Is this assertion true? or can you sequence fragments over 1200 bases?

[microgirl123]

Larger fragments don't hybridize to the MiSeq flow cell as well as smaller ones. This can be seen when you have a small amount of adapter dimer in your (larger) libraries. The adapter dimer binds preferentially and what started out as a small percentage of adapter dimer ends up being half your sequencing data! However, the main problem with large fragments is that they tend to "flop over" into other clusters when they're bridge-amplifying so the MiSeq can't differentiate between clusters. Why do you want such large fragments?

[Peter Meintjes]

Thank you - yes, I'm familiar with the bias towards smaller fragments, but am hoping to identify if there is an upper limit.

The reason for want to sequence larger fragments is that for targeted sequencing of the HLA region the further the distance between the paired reads, the greater the ability to phase the variants present between the two alleles in the sample.

Do you know if there is an upper empirical limit on what can be successfully sequenced on the MiSeq?

[GenoMax]

See this thread for related discussion: http://seqanswers.com/forums/showthread.php?t=20839

[nucacidhunter]

Following paper has described sequencing large amplicons on MiSeq:

http://biorxiv.org/content/early/2014/11/06/010967

Materials and Methods:
http://biorxiv.org/content/biorxiv/s...1/010967-1.pdf

[pmiguel]

Hi Peter,
To a first approximation, the issue is only one of competition from shorter amplicons. If you can completely rid yourself of all shorter amplicons, then the longer ones can cluster.
Generally everyone wants to blame the failure of the long amplicons to sequence on their supposedly creating large diffuse clusters. But I think this is a minor issue because I didn't see a dramatic change in the diameter of clusters between 500 and 1000 bp amplicons. (See the thread linked to by GenoMax above.)

--
Phillip