Natural compounds have been commercially used in pharmaceutical industry due to their therapeutic nature. Herein, natural compounds are screened that modulate various biological processes that are known to be conserved from yeast to human and to be associated with human diseases such as cancers and viral infections. Baker’s yeast Saccharomyces cerevisiae is a eukaryotic model in the pharmaceutical industry and its use allows genome-wide recognition of potential targets of natural compounds. Cancer is one of the leading causes of death worldwide with the number of diagnosed cases of this disease escalating. Chemotherapy remains the number one treatment option, especially in developing countries where other forms of treatment remain costly. The anthracyclines (natural compounds) are chemotherapeutic agents that have been in use for the last forty years and that are effective in treating various solid tumours as well as haematological malignancies. Various modes of action have been suggested for these agents; however, it remains unclear as to how these agents really exert their cytotoxic effects on cells. Anthracyclines can cause severe adverse effects such as cardiotoxicity as well as myelosuppression, and these are the major limitations of these drugs. To solve these problems, the controversy regarding their mechanism of action needs to be addressed. For this reason, here the focus is on the analysis of molecular mechanisms responsible for anthracycline cytotoxicity and the identity of the intracellular compartments that anthracyclines localise to in a mammalian cell line and relate the findings to those controversies found in the literature. In addition, natural compound-based potential anti-viral compounds that may inhibit cellular vacuolar protein sorting-4 (Vps4p) ATPase activity and in parallel block viral budding from the cellular membrane were also investigated. Natural compounds have been largely excluded from characterisation via high-throughput profiling strategies due to their limited abundance. A previous investigation used high-throughput yeast chemical genomic (CG) interaction profiling to permit the identification of putative cellular targets for the naturally-derived anthracycline-related compound epsilon rhodomycinone (ε-Rh) and the related synthetic compound 1,6,11-trihydrotetracene-5,12-dione (D11) in yeast. A set of 54 putative targets were identified using the wild-type yeast strain BY4741 and genome-wide set of congenic gene knockout mutants harbouring complete deletions of individual nonessential genes. Here, the focus is on the hypersensitivity of a specific subset of those mutant strains that each harbour a complete deletion of a different gene. Examining these hypersensitive strains can identify the possible targets whose altered activity is the molecular mechanism for anthracyclines. The results from this study identify and confirm the hypersensitivity of the BY4741 sod1Δ gene deletion mutant to the compound ε-Rh, D11 and clinically used anthracyclines (doxorubicin and daunorubicin). The identification of the sod1Δ gene supports the validity of this yeast-based approach for identifying cellular target(s) of these drugs, because anthracyclines have been reported to damage mammalian cells by generating reactive oxygen species such as superoxide and the SOD1 gene encodes superoxide dismutase, which is an enzyme responsible for breakdown of superoxide. Therefore, yeast haploid gene deletion mutants in which the gene encoding Sod1p is deleted (sod1Δ) display hypersensitivity to these drugs and this result agrees with expectations. A cultured mammalian cell line was also used to identify cellular compartments in which internalised natural and clinically used anthracyclines accumulate. The results from this study demonstrate that both ε-Rh and clinically used anthracyclines (doxorubicin and daunorubicin) localise within the nucleus. This study identifies that the main mode of action for these anthracyclines is in the nucleus with no functional interaction with plasma membrane or any other cellular compartment. These findings are in accordance with the previous collected yeast chemical genomics analysis data that showed radiation sensitive RAD genes (localised within nucleus) as cellular targets of these anthracyclines. Moreover, the disruption of SOD1 gene generates reactive oxygen-species that lead to the DNA double-strand breaks within the nucleus. Therefore, ε-Rh mode of action can be considered as a potential anti-cancer, but its toxicity needs to be tested. In parallel, the screening of a natural extracts (335 marine compounds) library using yeast as a model organism was completed with the aim of identifying a novel Vps4p inhibitor with potential application as an anti-viral compound. This exhaustive screening was not able to provide any positive hit; however, this study found that the commercially available DBeQ (an inhibitor of p96) is also an inhibitor of Vps4p in S. cerevisiae but did not abolish the Vps4p physical interaction with the associated protein Vps2p. Overall, the findings presented in this thesis has advanced the knowledge regarding the application of yeast for the screening of natural products to characterise the mode of action of bioactive natural products on all eukaryotic cells. This knowledge can then be used for the design of novel compound for cancer and/or enveloped virus infections.