How does Rcistarget utilize two databases for scoring motifs?

[KoichiHashikawa]

Hello,

I tried to understand how Rcistarget scores motifs by utilizing the two databases, for example, mm9-500bp-upstream-7species.mc9nr.feather and mm9-tss-centered-10kb-7species.mc9nr.feather.

Is Rcistarget searching for motif enrichment in 500bp upstream of TSS or 10kb +-?
Are motifs in the -500bp more valued than the ones found in 10kb+-?

If I want to expand the search window (e.g. 15kb), how I could generate the database?

In addition, in https://resources.aertslab.org/cistarget/databases/mus_musculus/, there are mm8,9,10. what are the major diffferences?

Thank you so much.