The mycological characteristics of fungi causing infections in humans, and the clinical features of the infections they cause, present unique parasite-host interactions that make them valuable research models to study. On the basis of biochemical analyses, antifungal drugs currently in clinical use have been developed for their capacity to inhibit enzymatic activities present in fungi but not in the host. Nowadays, the treatment of systemic mycoses depends largely on three classes of structurally related compounds: polyene antibiotics, flucytosine and azoles. The antifungal activity of polyene antibiotics, amphotericin B and nystatin, is due to the binding to sterols present in fungal cell membranes. Flucytosine, a synthetic nucleoside, by converting itself into 5-fluorouracil, interferes with fungal protein synthesis. Finally, the synthetic imidazole antifungals presumably act by inhibiting cytochrome P-450 sterol demethylase. However, these drugs have drawbacks associated with their usage, such as high toxicity for the kidney, high mortality rate (amphotericin B), and other problems associated with the use of flucytosine, such as resistance developing during therapy. Azoles usually have low toxicity, are fungistatic, but their use is not recommended for long-term treatment such as that required in AIDS patients. While further studies are being performed on some of these compounds to have a better understanding of their mode of action, there is also an urgent need for the development of new classes of antifungal drugs. Recent studies of known antifungals, together with new approaches based on the latest advances in molecular biology aimed at the identifcation of new targets that represent the basis for the synthesis of novel classes of fungal antibiotics, are reported. The principal objective of this approach is to identify and clone specific genes that are absent or whose products have low amino acid homology with corresponding human proteins, to selectively inhibit a fungal cellular function. However, this approach will rely not only on a better understanding of the mode of infection of pathogenic fungi, but also on the significance of 'virulence factors' which have been attributed to activities that, in the past few years, have been associated with some specific fungal gene products. Most of the studies performed to date on host-parasite relationships have been based on the elucidation of the mechanisms used by the human immune system to circumvent fungal infection, but too little is known of the molecular processes used by the fungus to invade a human host and establish infection, both in normal as well as in immunocompromised patients. This comprehension is definitely needed in order to