Large Library On Multiplexed Run

[samdini]

Hey all,

I'm currently running highly multiplexed runs on a Nextseq 550 and I've come into the problem where a library submitted to me has an average 950bp length and has been shown to cluster extremely inefficiently (i.e. only 10% of the reads loaded for QC sequencing were returned pass filtered). The clients are now requesting about 40% of a mid output runs worth of data to account for the inefficiency of their library (they expect to get back only 10% of the data again). I'm worried about the downstream effects of running this inefficient pool on the run as a whole, since many other collaborators and data from our own lab will be on this run.

[itstrieu]

I have sequenced large libraries before and while they do cluster inefficiently, the reads that passed filter are similar to other libraries. Mixing libraries of different sizes can be tricky in determining the ratio of each library to get the desired reads.