Would it be overkill to perform 12 samples for 16S sequencing using Illumina v3 kit?

[annaatobe]

Hi, I'm a very very new beginner in sequencing field. I'm getting microbial gDNA from environment samples and am thinking of doing Illumina 16S rDNA sequencing (metagenomics and RNA-seq in the future). Currently there's nobody in my lab doing sequencing with complex microorganisms sample and my boss want me to prepare the library by myself, so I'm just thinking of asking some stupid questions here for help...
Currently I have 12 gDNA samples (ranging from ~10 ng to ~1000 ng) and just want to do a 16S sequencing test run to get some data. I followed the 16S protocol in Illumina's website and target at the V3-V4 region (~550 bp totally). In order to get an overlap in the final sequence data we have booked the MiSeq reagent V3 kit (I think with its 2x300 cycle this is the only kit that can provide us with an overlap) and want to perform the sequencing on a MiSeq. However it's said that v3 kit can generate 13.2-15 GB data and is used for hundreds of 16S samples, in this case I'm kind of wasting it with only 12 samples. Could I ask for some suggestions on this issue?
Also I'm thinking if I'm doing metagenomics in the future maybe in this case I can also perform a metagenomics sequencing together with the 16S sequencing, using this v3 kit. Since I'm now only doing a test run, I don't need the depth and coverage to be that much. Would you think this might be possible?
Thanks a million in advance!!!!!!

[nangel]

Yes, using this kit for 12 16S amplicon samples will be a massive overkill. Even with 192 samples per run you get far more reads than you really need. If you have metagenomic samples that you were anticipating using for shotgun sequencing you could combine them and it would be useful to also increase your diversity and eliminate the need for PhiX in the run.

[annaatobe]

Yes, using this kit for 12 16S amplicon samples will be a massive overkill. Even with 192 samples per run you get far more reads than you really need. If you have metagenomic samples that you were anticipating using for shotgun sequencing you could combine them and it would be useful to also increase your diversity and eliminate the need for PhiX in the run.

Thank you very much for the reply! I just know that PhiX is used as a 'control' in 16S sequencing, would it be OK for me to ask if I combine 16S and metagenomics library together there's no need to use the PhiX control? Also in this case, since 16S might need much less data than metagenomics, may I ask should I combine the two library in half or should I put more metagenomics library in this library combination set? Thanks a million!

[nangel]

If you were just running the 12 16S samples then you would need to use PhiX to provide enough diversity in the sample (usually 10% or higher). Using your metagenomic library will provide that diversity for you so you would only need to use 1% PhiX as a control which means that you actually get more use from your reads. It depends on your expected community complexity for the 16S amplicon samples as to how much you might like to load them at. For most of our 16S samples we aim to have about 0.5% of the reads per sample from the run. This usually equates to at least 30 000 reads for the sample without any problems. If you wanted to balance the run I would say that you would be safe using 10% 16S pool to 89% metagenomic pool with 1% PhiX- obviously at the same nM And I guess I am assuming that your metagenomics library is to be at around the same fragment size as your amplicon size.