I've just wondered: why not just all adapters on one big plate? I'm a bioinformatician, so I apologize in advance if the answer is obvious to wet lab scientists.
Lanes are physical/real for MiniSeq/MiSeq (one lane), HiSeq (2 or 8 lanes) platforms. In case of NextSeq and NovaSeq, even though the lanes are optically distinct, they are not fluidically separate (same pool runs across all lanes) . In case of NovaSeq, use of an external manifold allows lanes on flowcells to be used as physically separate lanes (XP protocol). Same adapters are present on entire flowcell that allows for binding of library molecules.
As already indicated in a separate comment, physical lanes allow pools with different samples to share the same index sequences on same flowcell.
There is a limit to how many validated 8-mer or 10-mer indices one has, what if someone wants to run more samples than that at a time? A Hiseq flowcell I just processed has about 5000 samples on it, not all from the same submitter. That would be much harder to arrange without lanes.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
if you have more than one sample, more than one customer, it avoids cross-contaminations.
It also allows you to re-use indexes across lanes, which makes library prep a lot less annoying when you have a lot of samples.
Or to use different index strategies across lanes, for when users are making libraries themselves (Illumina doesn't support this, but it works fine).