Best way to generate reads for testing of aligners/snp callers

[gavin.oliver]

Hi all,

I know of numerous programs for generating artificial reads e.g. wgsim, bgeneratereads, simNGS...

BUT

I am trying to find one which will annotate the output sequences with a description of various characteristics that will make downstream testing of aligners/snp callers easier.

For example, the read name would contain info on chromosome and position of origin and the coordinates of any errors/snps/indels etc that have been introduced to the sequence.

Does anything like this exist or would it be necessary to write a program/hack an existing one?

Thanks in advance.

[nilshomer]

Shameless plug: http://sourceforge.net/apps/mediawik...ome_Simulation

[rskr]

Shameless plug: http://sourceforge.net/apps/mediawik...ome_Simulation

We are eagerly awaiting your evaluation scripts.

[nilshomer]

They are there, but not documented on the wiki. Maybe you could try it out and help add some documentation? Open source FTW?

[gavin.oliver]

Shameless plug: http://sourceforge.net/apps/mediawik...ome_Simulation

I was hoping you could answer this one

[elizabeth314]

I know this thread is old, but I thought I might add that you can check out DNemulator at

http://www.cbrc.jp/dnemulator/

It allows you to generate simulated reads from a reference, using the fastq scores of a run from your favourite platform to set the scores on your simulated data. It also saves the "mapping" of where your reads come from, so you can figure out whether your mapper is doing a good job or not...

Cheers,

Elizabeth