Aligning ChIP-seq reads to repeats: IgG vs input controls

Question

[jotan]

I would like to check if I have ChIP-seq reads that map to repeats in the genome (e.g. LINEs, SINEs, telomeres, centromeres etc.) I use the Kharchenko lab repeat aligner to do this.

I intuitively feel as if an Input sequencing control is the best control since this should give an idea of the copy number of repeats within a single genome. However, I have noticed that some people use an IgG control instead.

Are there any thoughts on which control is better for alignments to repeats specifically?