Single or Paired Read

[suludana]

Hi,
I will sequence a gene and I need to decide betwen single or paired sequencing.
What kind of criterion should I take? ... I know I must use paired-end if I want to read a G rich sequence or if I work with large rearrangements or delections... What else? Any other important thing?
Thanks

[apfejes]

I'm not sure I'm understanding correctly - are you planning on sequencing only one gene?

If that's the case, this is clearly not the right technology to use.

[suludana]

I'm not sure I'm understanding correctly - are you planning on sequencing only one gene?

If that's the case, this is clearly not the right technology to use.

Hi,
yes, I'm planning on sequencing only one gene, but in many samples.
¿Why do you think is not the right technology?

[apfejes]

If you're doing one gene, but many samples, then the only way to get any value out of using a massively parallel sequencing reaction is to do barcoding. With that said, barcoding has it's issues, and - unless you're doing 1000's of samples - will probably give you excess coverage, and be a pain to re-assemble.

As far as I can tell, no one has come up with a convincing/compelling reason to do a single gene using Illumina/454 technology. The depth will be excessive, the cost will probably exceed that of doing a normal Sanger reaction, and the individual reads won't be long enough to cover anything but the shortest genes.

Did you have something else in mind?

[suludana]

If you're doing one gene, but many samples, then the only way to get any value out of using a massively parallel sequencing reaction is to do barcoding. With that said, barcoding has it's issues, and - unless you're doing 1000's of samples - will probably give you excess coverage, and be a pain to re-assemble.

As far as I can tell, no one has come up with a convincing/compelling reason to do a single gene using Illumina/454 technology. The depth will be excessive, the cost will probably exceed that of doing a normal Sanger reaction, and the individual reads won't be long enough to cover anything but the shortest genes.

Did you have something else in mind?

Yes, of course I’m thinking about barcoding. If I work with a relatively big amount of samples the costs may be lower than if I do a normal Sanger reaction.

I read some papers about sequencing 50-120 kb fragments using Illumina/454 technology. They made multiplexing with 30-50 samples.

The posterior data analysis will be complex, but I think it will be possible. It´s not easy to profit this technology. Undoubtedly the preparation of the samples for subsequent sequencing is the more complex step. Amplify long regions by PCR, or make capture by array, always comes out expensive.

[Melissa]

I think the question is

• Do you have a GA/GAII? If you don't, barcoding is very expensive because you'll be charged library construction for each sample you have.

• What's the size of your gene? If your gene is a few kb, then forget about it.

• How is your gene sequences? You have to look at the sequences and decide whether short reads are suitable. If your gene contains region of repeat or high similarity, the reads will not be mapped correctly. Paired-end sequencing will definitely help in mapping. And let me take a wild guess, your purpose of sequencing mulitple samples is to find SNPs, is that right?

The cost is definitely much lower than Sanger sequencing and it gives deep coverage. Both require PCR amplification. Library construction for Solexa multiplex sequencing is laborious. Assembly of short reads can be a problem. I have to agree with Anthony that the technology is not mature yet.