Conceptual question about integration

[dgagler]

Hi all, thanks for the great package. A question about integration:

I have a set of scRNA data that includes some trusted annotations on various B cells. I have scATAC data that is justifiably expected to include a broader immune set including B cells, T cells, monocytes, etc. Can I still use this scRNA to integrate or should I find a scRNA reference that includes a broader immune population? I'm actually only interested in the B cells in my scATAC data, but I can't remove them until I've annotated them and I'm concerned that addGeneIntegrationMatrix() will forcibly assign B cell labels to T cells/monocytes and I will therefore not know to remove them from downstream analysis. Thanks!