This site has been established to measure structural and functional distinction between sequences of proteins. This approach is tackled by adapting a statistical approach using mutual information (MI), a measure of association to reveal group-specific characteristics of conserved sequence motifs. In this environment group-specificity may be understood as: specificity by species, by phylogeny, by position or by any other user definition. The current stage of this service may be seen as a proof of concept, rather than a high performance high throughput tool.
For more information, you can also see the paper on PROMI in BMC Bioinformatics entiteld "Species-specific analysis of protein sequence motifs using mutual information".