Simple questions about Illumina

[Liselotte]

Hi,

I am currently working on my thesis and I have just a few questions. We want to sequence a whole transcriptome using 2 x 150 bp read lenghts on the illumina. (the fragments to be sequenced are about 170 bp long). But why is this paired end better than single end ?

In my opinion, paired end reads from the 2 sides of the fragment you're sequencing. But is the second read performed on the same fragment or the complementary fragment ?

Finally, I'd like to know the advantages of this Illumina when you compare it to Roche or Solid.

These questions may sound silly, but I'm a complete noob at next-generation sequencing. So please try to simplify your answers

Thank you already

[Chudasama]

Hi,

First of all, single reads gives you the sequence information.

Paired end reads gives you the sequence information and it provides you a with the knowledge that these 2 sequences are "grouped", i.e. if you map one of the sequences, you know that the other one should be in close proximity (as long as your fragments are short 170 bp, 170bp - 72 bp from each side gives a space of 36 bp in between). So the sequences should be placed on the genome with around 36 bp from each-other...

With illumina the fragments are hybridised to the flow cell, bridge amplifies (synthesize of the complementary sequence, again linearized and then sequenced in reverse). So its the complementary strand which is the template for the reverse sequence.

All the platforms have advantages and disadvantages. illumina operates with shorter reads than the 454, but have more reads then the 454. Search for illumina, solid and roche comparison, and you will find several links.

C

[Liselotte]

Allright, thanks a lot!

[HESmith]

Just curious; if your fragments are only 170bp, why would you want to run 2 x150bp? Most of the data will be redundant b/c of overlapping reads.

[Liselotte]

Good question and it appears I was wrong. We will be performing a single read of 150 bp...

[kmcarr]

Good question and it appears I was wrong. We will be performing a single read of 150 bp...

If it were me I would prefer to do paired end reads, 2x75, in this case instead of a single end 1x150. The reason is the drop in quality scores for the longer reads. With the resynthesis step in the between the reads the quality pops back up to (nearly) the same level it was at the start of read 1. You still get 150bp of data for each read but the quality (accuracy) of the reads will be better. There will be only a marginal increase in cost for the paired end vs. single end flow cell and cluster kit.

[HESmith]

I second kmcarr's suggestion; you'll get much more high quality data w/ 2x75 reads.

[Chudasama]

I agree, I thought thats what you meant, so 72-75bp reads from each side...