Biological species collections are critical for natural product drug discovery programs. However, prioritization of target species in massive collections remains difficult. Here, we introduce an untargeted metabolomics-based prioritization workflow that uses MS/MS molecular networking to estimate scaffold-level distribution. As a demonstration, we applied the workflow to 40 polyporoid fungal species. Nine species were prioritized as candidates based on the chemical structural and compositional similarity (CSCS) metric. Most of the selected species showed relatively higher richness and uniqueness of metabolites than those of the others. Cryptoporus volvatus, one of the prioritized species, was investigated further. The chemical profiles of the extracts of C. volvatus culture and fruiting bodies were compared, and it was shown that derivative-level diversity was higher in the fruiting bodies; meanwhile, scaffold-level diversity was similar. This showed that the compounds found from a cultured fungus can also be isolated in wild mushrooms. Targeted isolation of the fruiting body extract yielded three unknown (1-3) and six known (4-9) cryptoporic acid derivatives, which are drimane-type sesquiterpenes with isocitric acid moieties that have been reported in this species. Cryptoporic acid T (1) is a trimeric cryptoporic acid reported for the first time. Compounds 2 and 5 exhibited cytotoxicity against HCT-116 cell lines with IC50 values of 4.3 and 3.6 μM, respectively.