Pooling across lanes and depth

[darthsequencer]

Hi All,
I'm sequencing ~36 metagenomes and I'm thinking about how to get the best coverage.

The easiest thing would be to pool all 36 and run them across lanes - this will also help with lane to lane variation. But I think this will reduce the overall sequencing depth of each sample (index). As I'll just be resequencing the more abundant sequences (organisms) over and over.

So will it actually help to run say 8 samples per lane to make sure that the same clusters are not being sequenced again and again.

Thanks

[dpryan]

There is no down-side to splitting each sample over all of the lanes (aside from the fact that you need more barcodes...but 36 is doable), your sequencing depth due to this is unaffected. The index is a separate read entirely, so it's not like having it there decreases your read length or cluster count.

[LOH]

Sequencing is counted by million reads per lane. I don't think there is any final sequencing depth differences if final sequencing lanes are the same. It would be better if you pool all 36 and run across lanes, just make sure ---balance each library before you mix them together.