Before I get to your question I just wanted to let you know that UniProt (http://www.uniprot.org/uniprot/O75594#ref6
) is also a great web resource.
So the premise of "Computational Exploration of Protein Function" is that if some protein is really important (i.e. required for basic cellular function) it probably is not changing much over evolutionary time. A non protein example would be 16S ribosomal RNA. It is very important for translation and has many conserved regions within gene sequence and can be used to identify microbes across the two domains of prokaryotes (http://en.wikipedia.org/wiki/16S_ribosomal_RNA
I know that this example seems tangental but if you looking at the clustal alignment of PGLYRP1 are there regions that are different across species or that remain the same. Also remember that there are different groups of amino acids and that they form very different 3-D structures (http://en.wikipedia.org/wiki/Secondary_ ... _structure
). A change from an hydrophobic amino acid to another hydrophobic amino acid with change protein structure less than a change from an hydrophilic amino acid to an hydrophobic amino acid.