Alignment of samples with spike-in

Question

[maria.soler]

Hi, I have a question about spike-in normalization. Do you use a custom reference genome combining your organism and the one used as a spike-in? I think it's a common approach, but then if you only allow to map uniquely mapped reads, you are losing spike-in information, aren't you? Since spike-in mapping is only for quantitative purposes, I think it would be better to align to both genomes separately, allowing multimapping for spike-in, and then filter BAM files to exclude reads present in both of them. But I am not sure if I am missing anything, so any input or advice is welcome.

Thanks a lot in advance!

Eugenia

[GenoMax]

but then if you only allow to map uniquely mapped reads, you are losing spike-in information,

How so? Spike-in's are generally unique sequences that don't occur in your sample of interest. Otherwise you would not be able to recognize the spike-ins.