for example, right now it looks like a nf pipeline is being used to build some ref files, which I assume would be ones that are not available publicly like the BWA indexes, etc.. I was not clear though what the intended relation was between this and the pre-existing public datasets is, such as builds and annotation sets from UCSC and other institutions, and also the relation with the current AWS igenomes S3 repo and its files and genome build release bundles is this going to be a full replacement of igenomes? Or is this just for the non-igenomes files? Or both?