Identifying all pairs of inverted repeats

[Jon Lerga Jaso]

Hello,

I want to download those sequences that are repeated in the genome, and are similar. Concretely, all pairs of inverted repeats.

For example, I can download Segmental Duplications (hg19, Repeats, Segmental Dups), and it is very clear that, in the same row, I can see the sequence position of the first duplication and the position of the second duplication, their similarity and their orientation (+/-).

I think we have the same schema for chained self aligments (Self Chain), with the position of the first sequence and the similar sequence.

However, I have two questions:

* First, if the self aligments are those alignments of the human genome with itself; then, do they include the information of the segmental duplication track?

* Secondly, if I should download segmental duplications and self-alignments in order to achieve my goal, should I also download the repeat-masked sequences (RepeatMasker)? The problem here, it is that repeats are not associated with another sequence in a clear way. Do we have to blast each sequence with the others? Is there some way to take only those pairs of repeats that are inverted?

Thanks for the time!

Jon