I spent quite a bit of time on the Urton database in 2006. Would need to dig out the pre/post processing Perl scripts and see a more recent version of the database.

An area that would also be ripe for computational analysis is the texts from Spanish archives that are known to have been sourced from khipu readings. "Pärssinen, Martti & Kiviharju, Jukka. Textos Andinos. Corpus de textos khipu incaicos y coloniales. Tomo I & II" would be a good starting point.