16s primer design for Illumina platform

[urchin]

I am another newbie trying to get through a first project in NGS.

We are trying to amplify universal prokaryotic sequences for 16s rRNA gene (V3-V4 region) from benthic marine critters - mostly from microbiome extractions using materials from invertebrates and some algae. I was told that using the Fadrosh et al. 2014 dual-indexing method would improve our results a lot when we do PE250. Hopefully that is true - yes? (No?)

Also, right now all the current primer sets are going through continued evaluation and most seem to be found lacking in one way or another. Does anyone have a strong opinion on choices based on actual experience with getting good results? Has anyone tried the primers discussed in Takahashi et al. 2014, Pro341F/Pro805R? Did they work as well as claimed?

Thanks for any help pointing me in the right direction!

[jtotheulie]

Hi,

Do you have an update on this project? Did you test Fadrosh et al. 2014 dual-indexing method?

thanks

[urchin]

Hello, No I did not. I am using the Kozich method. I was able to get in touch with a friend's colleague who is doing this type of work already and he informed me that the problems with libraries with low diversity getting good reads with Illumina Miseq for PE250 have been resolved with new chemistry. He said the additional modifications really weren't necessary anymore.

Anyone have another opinion about that??

[nucacidhunter]

We are trying to amplify universal prokaryotic sequences for 16s rRNA gene (V3-V4 region) from benthic marine critters - mostly from microbiome extractions using materials from invertebrates and some algae. I was told that using the Fadrosh et al. 2014 dual-indexing method would improve our results a lot when we do PE250. Hopefully that is true - yes? (No?)

Thanks for any help pointing me in the right direction!

For optimum quality 16S libraries has to be spiked in with high diversity libraries and clustered in lower density regardless of improved chemistry. Adding heterogeneity spacer will enable increasing cluster numbers therefore maximising sequencing yield without adverse effect on quality. The issues with the method are:

1- Adding inline index decreases useful read length. This is OK for short target regions but unsuitable for longer targets such as V1-V3.
2- There are some possible issues with reproducibility using different spacers that have not been addressed by the authors (they have not used a known control community with all of their primers to see if they get the same results).
3- Long primers are less efficient and more biased in amplification.

[urchin]

The Illumina sequencing site we are sending our library to will use Phix to spike the library. I'm guessing at around 30%?? Is that adequate to deal with this?

I'm not clear if you are really trying to recommend the spacers or not with your additional points it seems they could be a problem for us.

We are planning on amplifying V3-V4 using 341f and 785r, not quite as long as V1-V3 so would the inline index present a problem for our target length getting read adequately?

[nucacidhunter]

The Illumina sequencing site we are sending our library to will use Phix to spike the library. I'm guessing at around 30%?? Is that adequate to deal with this?

I'm not clear if you are really trying to recommend the spacers or not with your additional points it seems they could be a problem for us.

We are planning on amplifying V3-V4 using 341f and 785r, not quite as long as V1-V3 so would the inline index present a problem for our target length getting read adequately?

20% PhiX with cluster density below 600 would be fine.

Generally spacers are good and one can increase cluster density and reduce PhiX spike in (more reads). The issue is that one have to validate their design to make sure that spacer does not cause any biases. One way to do this would be using control mock community with known species and amplifying target V region followed by sequencing and analysis. If there is no bias all primers should give very similar quantitative and qualitative results. The paper has not mention the validation.

[jtotheulie]

Has anyone modified the 16s illumina protocol for other amplicons. I want to use it for 16s, its, and 18s. If we can do this then we can use the same sequencing primers and sequence them in the same run. Avoiding the conserved sequencing issue that illumina has and the confusion of the EMB protocol that uses different sequencing primers between the 16 and 18s.

Any thoughts?

[thermophile]

I've modified Kozich primers for ITS and family specific portions of 16s. Doing this requires spiking in a sequencing primer for each target (the sequencing primer is the pcr primer plus a pad). This works well. If you're running lots of samples with a particular target, I'd recommend this method since you don't waste sequencing cycles on your pcr primers.

But for those cases where you aren't running lots of samples on a target, I'm also doing a 2 step target agnostic amplicon design based on http://bmcgenomics.biomedcentral.com...471-2164-15-63

[urchin]

phasing amplicon approach - alternate spacers?

20% PhiX with cluster density below 600 would be fine.

Generally spacers are good and one can increase cluster density and reduce PhiX spike in (more reads). The issue is that one have to validate their design to make sure that spacer does not cause any biases. One way to do this would be using control mock community with known species and amplifying target V region followed by sequencing and analysis. If there is no bias all primers should give very similar quantitative and qualitative results. The paper has not mention the validation.

The Fadrosh work has now been followed up with a modification on the spacers by Wu et al. 2015. Do you think that the control tests they did are adequate? I am now considering using this method for our primer spacers instead of just depending on the chemistry at Illumina being good enough for the lack of diversity. This method will not cause us to lose so much of our sequencing length, as well.

By the way, does anyone think there is a problem with using PE300 instead of PE250 with this method????

[nucacidhunter]

Could you give more details for Wu et al 2015 to clarify the paper you are referring.

[urchin]

Could you give more details for Wu et al 2015 to clarify the paper you are referring.

Wu et al. BMC Microbiology (2015) 15:125
DOI 10.1186/s12866-015-0450-4

Phasing amplicon sequencing on Illumina Miseq for
robust environmental microbial community analysis
Liyou Wu1†, Chongqing Wen1,3†, Yujia Qin1†, Huaqun Yin1,4,5, Qichao Tu1, Joy D. Van Nostrand1, Tong Yuan1,
Menting Yuan1, Ye Deng1,7 and Jizhong Zhou1,2,6*

They have spacers that always are 7 bases long, by altering the spacer insert length on 5' versus 3' primers. They did include various control tests, including against a mock community.

Here is a quote from the abstract: "Our results first indicated that the PAS method substantially ameliorated the problem of
unbalanced base composition. Second, the PAS method substantially improved the sequence read base quality
(an average of 10 % higher of bases above Q30). Third, the PAS method effectively increased raw sequence
throughput (~15 % more raw reads). In addition, the PAS method significantly increased effective reads (9–47 %)
and the effective read sequence length (16–96 more bases) after quality trim at Q30 with window 5. In addition,
the PAS method reduced half of the sequencing errors (0.54–1.1 % less). Finally, two-step PCR amplification of
the PAS method effectively ameliorated the amplification biases introduced by the long barcoded PCR primers."

[urchin]

Link to Wu et al 1015

http://bmcmicrobiol.biomedcentral.com/articles/10.1186/s12866-015-0450-4

[kerplunk412]

By the way, does anyone think there is a problem with using PE300 instead of PE250 with this method????

The 600 cycle v3 kits were showing really poor base qualities toward the end of reads. I am not sure if Illumina has resolved this yet.

[nucacidhunter]

@urchin

Wu's work seems has addressed the amplicon sequencing issues on Illumina platforms adequately. I would suggest a small trial if you are amplifying other regions or modifying their protocol with one control sample.

[urchin]

Thanks for the information, I thought it looked like they had done a relatively thorough job, but I am new to this area.

I wanted to do V3V4 but now am very concerned that I won't get enough sequence length to overlap for alignment if I have to use PE250 to get decent reads. Above "Kerplunk412" indicated the PE300 is not giving good quality reads at the ends, so I don't see how that would help me get more useable data. Does anyone have possible suggestions or knowledge about what specific sequence length I could get with PE300?