sequencing paired-end and mate-pair libraries in a single lane?

[Vinn]

Hi folks,

We are new to genome sequencing, but we want to sequence fungal genomes using Hiseq. We were wondering if it is possible to include both mate-pair (2 or 3 kb) and paired-end (either 300 or 550 bp) libraries in one lane?

And if not possible to do so, which type of libraries (paired-end/mate-pair) shall we go with? Our fungi have their genome size of ca. 60 MB. At the moment, we are accessible to 2x150 but it can be changed to 2x250.

Thank you in advance for all advice!

[nucacidhunter]

Multiplexing those libraries in one lane is fine as long as they have color balanced index.

[Markiyan]

Dear Vinn,

You can mix those provided they have compatible indexes, but watch out for differences in the fragment size distribution (and clustering efficiency) and use qPCR for pooling.

I assume you're running your's HiSeq in 2x250 bp mode.
And the libraries would be prepared using a PCR-Free truseq and Nextera Matepair kits.

Otherwise a single 2x250 or 2x300 MiSeq run would give you better results than a 2x125 HiSeq. (provided you get at least 30X coverage).

[kmcarr]

Originally Posted by Vinn

Hi folks,

We are new to genome sequencing, but we want to sequence fungal genomes using Hiseq. We were wondering if it is possible to include both mate-pair (2 or 3 kb) and paired-end (either 300 or 550 bp) libraries in one lane?

And if not possible to do so, which type of libraries (paired-end/mate-pair) shall we go with? Our fungi have their genome size of ca. 60 MB. At the moment, we are accessible to 2x150 but it can be changed to 2x250.

Thank you in advance for all advice!

I have done a couple of fungal genomes (40 Mbp) now mixing paired end and mate pair libraries in a single run with no problem. One MiSeq PE250 run will provide plenty of data (6-7 Gbp) for your project. We did a single paired end library with a 550 bp insert size and had that make up ~50% of the pool on the run. We did three different mate pair libraries, 2, 4 & 8 kbp and had those three combined make up the other 50% of the pool.

One thing to keep in mind when planning your experiment and library design is what tools/software you plan to use for the assembly. I used velvet & Gap Closer, but some de novo assembly software (e.g. ALLPATHS-LG) insist that the paired end reads overlap so the fragment size for that library must be < 2 x read length. If you plan to use ALLPATHS (or another assembler with the same restriction) then you should adjust the size of your paired end library to ~350-400 bp if you plan to use PE250 reads.

[Vinn]

Hi folks

Thank you again for all great and useful advice. Just a small question, do you think in my case it is possible to get a good assembly with only a 550bp paired-end library? Or it comes again down to a matter of coverage?

[Markiyan]

Dear Vinn,

You can try doing shotgun only assembly, but yours results would depend on level of repetitiveness/ploidy of your species.

It probably would be sort of ok for most ascomycetes, but I would not dare for good results from basidiomycetes (unless you prep DNA from haploid mycelium).

PS: It is never late to do another matepair run and reassemble...