I'm trying to find pathways enriched by a list of genes, by variants per kb. Everything I read and see about IPA says it's for transcription data. I have whole exome data. Can I do this? Under Core Analysis, the data classifications are either Variant loss/gain, Variant ACMG Classification, or various expression labels.
IPA only provides pathway over-representation analysis. And for sure, you can upload any kind of data, including the gene's mutation load (variants per kb). In core analysis, you need to filter your data set by some cut-off, e.g., more than 1 variant per kb.
But keep in mind, that advanced analytical features of IPA like Upstream Regular Analysis are meaningless in this setting and won't work. This is because they need fold-change information which you can not provide.
However, IPA is probably overkill for your application. Because you can already do the filtering for "significantly/highly mutated genes" (e.g., >1 variant/kb) in Excel and the submit the list to any software (e.g, DAVID or EnrichR).
The filtering for "significantly/highly mutated genes" is crucial and a filter criteria is hard to find and even harder to evaluate (also keep in mind the different strength of effects). In your case, I would rather prefer to to a Set Enrichment Analysis that does not require a cut-off. Just use the mutation load as "association with phenotype" (or fold-change) in any GSEA tool you like.