The polyoma virus genome is organized in such a way that the DNA coding capacity of the virus is maximized. Not only are there overlapping genes, but more than one reading frame within a particular region of the genome is used for coding. This is especially apparent in the ‘ early region’, which codes for three of the known viral proteins, the large, middle and small T-antigens. Only a part of this region appears to be necessary for the transformation of cells, and it contains coding information for middle and small T-antigens, but only for a part of large T-antigen. A two-stage model for transformation is proposed which takes into account the major assays for transformation and the viral proteins that are suggested to be involved at each stage. In addition, the virus may induce a host protein to transform by an alternative route. It is suggested that the viral transforming function in polyoma virus could be of host cell origin.
