non-equimolar pooling to normalise read depth in libraries with different genome size

[kerryp]

I'd like to sequence libraries of very different genome size on a single MiSeq run.

Eight bacterial samples with 4.5Mb genome size
One eukaryote sample with 23Mb genome size

Could I pool the libraries so that the larger eukaryotic sample represents 38% of the pool, rather than 11%? In theory, this would allow read depth to be the same across all nine libraries.
But what effect would it have on the MiSeq sequencing process if one barcode combination is present at x5 greater levels than the other eight?

Has anyone tried to do this, and did it work?

Thanks in advance

[thermophile]

The only concern is index diversity, Check your combinations to make sure that all channels are roughly equally represented across the indexes given that one index will be ~40% of the reads.

[JakobHedegaard]

Hi kerryp,
No problem. Have done it successfully for several years doing Nextera XT libraries of bacteria and sequencing on NextSeq medium in batches of 50-70 bacteria.
I havent had concerns about the indexes but as pointed out by thermophile you must give it some thought running only 9 samples.
Good luck,
/Jakob