4C and MiSeq sequencing

[fabp]

I am carrying out circular chromosome conformation capture (4C), largely following the van de Werken protocol.

I've amplified my libraries using pairs of primers specific to different viewpoints. These are all sequence-specific primers with Illumina sequencing adapters on the 5' end.

I was pointed to these adapters by technical support on page 10 of the customer sequence letter - they are under TruSeq HT Sample Prep Kits. I was told by another contact that these are the "universal" adapters that should work on MiSeq and HiSeq. The sequences are below:

AATGATACGGCGACCACCGAGATCTACACACACTCTTTCCCTACACGACGCTCTTCCGATCT - read adapter
CAAGCAGAAGACGGCATACGAGATGTGACTGGAGTTCAGACGTGTGCTCTTCCGATC - nonread adapter

Amplification with the adapter-tagged primers appeared successful when running the products on a gel, and analysis on the Bioanalyzer indicated that I had a peak of a few hundred bp for each library (with some DNA at higher and lower sizes).

However, when I sequenced on the MiSeq, no clusters were detected. It doesn't seem to be a machine issue - I've been advised that it's probably the custom primers or the libraries.

I am going to do qPCR with NEBNext Library Quant Kit to see if the adapters are present (I believe this should work, even with my longer adapters?). But does anyone have any idea what might have done wrong here? Are my adapters wrong?

Thanks!

[fabp]

A related question about sequencing primers:

When I sequenced my amplified libraries on the MiSeq, I used my custom primer (adapter plus sequence-specific region) as one read primer. Since then I have wondered if I do in fact need to use a custom sequencing primer - since my amplified libraries have the adapters on, could I just use the standard Illumina primers in the cartridge?

What is normally done for 4C?

Thanks!

[simsalabim]

Hi,

I'm a little confused why you reverse complemented the sequence of the "nonread adapter".

just to make sure, here is the design as it should be put on the sequencer:

AATGATACGGCGACCACCGAGATCTACAC[i5]ACACTCTTTCCCTACACGACGCTCTTCCGATCT+YOUR SEQUENCE+GATCGGAAGAGCACACGTCTGAACTCCAGTCAC[i7]ATCTCGTATGCCGTCTTCTGCTTG
--------- binds to flow cell ----------index-------- sequencing primer---------------+YOUR SEQUENCE+-----------sequencing primer --------------index-------binds to flow cell------

Both adapters need to be attached in that direction. I know the 4C protocol is quite complex, make sure all your sequences are placed correctly. The way you posted the adapters, an issue might have been the wrong orientation of the second adapter? Then the clustering on the machine would fail.

To answer your second question:
Of course you can use the standard Illumina primers for the sequencing. But you would see your site specific 4C primer at the beginning of the read then. If you have more than one view point, you could use this information to assign the reads to the corresponding viewpoint though. On the downside, your read contains you primer then, so you sort of loose length. If you sequence 100/150 or so anyways, this should be affordable though.

Hope that helps...

[fabp]

Thank you simsalabim. I was told by the technical support at Illumina that the P7 primer needed to be reverse complemented before purchase so the sequence of P7 is in the correct orientation, but maybe I misunderstood. I will have a more in depth look at this to figure out what wrong.

However, the qPCR with NEBNext Library Quant Kit produced successful amplification, suggesting we have clusterable DNA. Would we expect this to work if the adapters are in the wrong orientation, or does qPCR success imply the sequencing should work?

Thank you for the information about the standard primers. We are doing 150bp read lengths so I think this should be fine!

Thanks again

[simsalabim]

Hey fabp,

so the sequence I wrote down is the one that needs to be present AFTER all your PCR steps => the one that will be put on the sequencer. So originally you start with the reverse complement of the P7 adapter, you're right.

P5 AATGATACGGCGACCACCGAGATCTACACACACTCTTTCCCTACACGACGCTCTTCCGATCT + specific primer
P7 CAAGCAGAAGACGGCATACGAGATGTGACTGGAGTTCAGACGTGTGCTCTTCCGATC+ reverse comp. of specific primer
These are your sequences then, right?

If you were able to quantify the library with the NEBNext Q Kit, this does suggest it is clusterable. Maybe there was an issue with the sequencing chemistry? Have you tried to cluster again or just once?

[fabp]

Thanks simsalabim. Yes, those are my sequences - the sequences of my primers and the ones that will be on my amplified DNA.

So as I was able to quantify, I will try to cluster again - I've only tried once so far. I will also use the standard Illumina primers - I'm hoping this will help, as apparently using custom primers can be less successful.

[kinvel]

I’m trying to put my question here because it has much in common with the above questions of fabp. I’m in the middle of similar 4C-seq experiment and now designing primers for final PCR. I did it according to van de Werken protocol (Meth. Enzymol., 2012, 513, 89-112) and my final library structure is expected to be as follows (1):

5’AATGATACGGCGACCACCGANNNPPPPPPPPPPPPPPAAGCTT<INSERT>PPPPPPPPPPPPPPPPPPPPATCTCGTATGCCGTCTTCTGCTTG

here are (from left) Illumina P5 primer, NNN – trinucleotide index to sort libraries (16 libraries should be sequenced in one run), PPPPP – viewpoint primers, and at 3’ end – the Illumina P7 primer, reverse complemented, length of each primer is about 43 n.

The libraries are planned to be sequenced (single end) using Illumina 2500. If I properly understand the technology, after bridge PCR and removing of P5-anchored fragments, the residual fragments attached to support through P7 will contain the reverse complemented P5 primer at their free ends and are sequenced from P5 primer.

However, when I discussed this experiment with our local NGS people, I was told that this library structure will not work and I will need custom sequencing primers, and proposed what they call universal adapters. With these adapters, the library structure (without viewpoint primers) should be as shown below (2):

5’-AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCTTCCGATCT-INSERT-AGATCGGAAGAGCACACGTCTGAACTCCAGTCACNNNNNNATCTCGTATGCCGTCTTCTGCTTG-3’

I don’t like this structure very much, because it has 60 n adapters, and with addition of my viewpoint primers they will become as long as ~80 n. To add these adapters it will be necessary to use nested PCR, and I would like to minimize the number of PCR cycles.

So my question is could the libraries with the structure (1) be sequenced using HiSeq 2500 with Illumina P5 primer, or it is necessary to use the universal adapters to prepare libraries with structure (2).

Thanks in advance.

[kinvel]

I’m trying to put my question here because it has much in common with the above question of fabp. I’m in the middle of similar 4C-seq experiment and now designing primers for final PCR. I did it according to van de Werken protocol (Meth. Enzymol., 2012, 513, 89-112) and my final library structure is expected to be as follows (1):

5’AATGATACGGCGACCACCGANNNPPPPPPPPPPPPPPAAGCTT<INSERT>PPPPPPPPPPPPPPPPPPPPATCTCGTATGCCGTCTTCTGCTTG

here are (from left) Illumina P5 primer, NNN – trinucleotide index to sort libraries (16 libraries should be sequenced in one run), PPPPP – viewpoint primers, and at 3’ end – the Illumina P7 primer, reverse complemented, length of each primer is about 43 n.

The libraries are planned to be sequenced (single end) using Illumina 2500. If I properly understand the technology, after bridge PCR and removing of P5-anchored fragments, the residual fragments attached to support through P7 will contain the reverse complemented P5 primer at their free ends and are sequenced from P5 primer.

However, when I discussed this experiment with our local NGS people, I was told that this library structure will not work and I will need custom sequencing primers, and proposed what they call universal adapters. With these adapters, the library structure (without viewpoint primers) should be as shown below (2):

5’-AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCTTCCGATCT-INSERT-AGATCGGAAGAGCACACGTCTGAACTCCAGTCACNNNNNNATCTCGTATGCCGTCTTCTGCTTG-3’

I don’t like this structure very much, because it has 60 n adapters, and with addition of my viewpoint primers they will become as long as ~80 n. To add these adapters it will be necessary to use nested PCR, and I would like to minimize the number of PCR cycles.

So my question is could the libraries with the structure (1) be sequenced using HiSeq 2500 with Illumina P5 primer, or it is necessary to use the universal adapters to prepare libraries with structure (2).

Thanks in advance.