Comparing molecular subtypes identified by microarray or bulk RNASeq in ssRNASeq data

Question

[akhattri]

Several studies have identified gene expression subtypes of different cancers. For example, this study has identified gene expression subtypes of head and neck cancer using microarrays. I would like to classify samples in this single-cell RNASeq data based upon the gene expression subtypes that were identified by microarrays.

I was thinking of converting single-cell RNASeq data to bulk RNASeq and then applying the classification but I am not sure if this can be done. The method of "pseudo-bulk analysis" cannot be used because it is used to compare clusters, but I want to identify subtypes initially before doing any comparison. If there is any tool or if this can done by any other approach?