questions about choosing read coverage

[christinawu2008]

Hi everyone,

It sounds stupid question... but I really want to know how you guys choose the read coverage before sequencing. I know it is related to the cost. But What is the minimum size and what is the best choice? and why? any rules here? (illumina)

Some useful link will be great..

Thank you!

[suludana]

Hi,

the necesary coverage depends on the proyect you are performing. For instance, if you are doing de novo sequencing probably you will need at least 30-50x, but if you are doing resequencing a 10-20x can be enough.

The way to calculate these values is:

The number of clusters you have per sample (read1 + read2 if it is a PE run) x read lenght / aproximate size of the sequence you are studing.

I hope this explanation will help you.

[Heisman]

On top of what suludana said, generally when aligning it isn't possible to align 100% of the reads. Maybe only 50-90% depending on the type of data that is being generated, so that should be factored in as well. If you want to remove low quality bases for whatever reason (for example, trim the last 30 bases of a 101 paired end read), then that needs to also be factored in.

[christinawu2008]

Hi,

the necesary coverage depends on the proyect you are performing. For instance, if you are doing de novo sequencing probably you will need at least 30-50x, but if you are doing resequencing a 10-20x can be enough.

The way to calculate these values is:

The number of clusters you have per sample (read1 + read2 if it is a PE run) x read lenght / aproximate size of the sequence you are studing.

I hope this explanation will help you.

Hi suludana,

Thank you!

Are there any literatures described how large is the expected read output (Gbases) for each sample in illumina RNA-Seq? (mammalian) if the genome is 3Gbases.

[trickytank]

We find that detecting heterozygous variates reliably is more difficult than detecting homozygous variants. If your samples are not from a consangous pedigree than I'd be aiming for 20x plus to avoid the false positives.