It is only now that, for the first time, scientists can bring progress in next generation sequencing and their knowledge and understanding of microbes, together with massive computational power and newer algorithms to accurately predict structures and functions of hundreds of thousands (or more) of proteins!
For decades, scientists have studied both proteins and their structures, as well as microbes and how they impact human health. However, those studies were greatly limited in terms of their scale (e.g. by studying one microbe at a time) and scope. Similarly, structures of individual proteins have been experimentally determined since 1958 and computational investigations of protein structures began in early 1970s.
A turning point came in the early 2000s with the introduction of next generation sequencing, due to progress in computing power and the development of new algorithms. Thanks to next generation sequencing, obtaining DNA sequences encoding genes became much cheaper and quicker.
Around the same time new tools, such as Rosetta (which is being used for this project), were being developed to computationally predict protein structure and were, in fact, used for the Human Proteome Folding project on World Community Grid.
Since then, these tools have been refined and enhanced. Combined with the massive computational resources of World Community Grid, a project of this scale has only now become possible.