A quantitative structure activity relationship (QSAR) modeling approach based on the location of each amino acid along three axes obtained by principal components analysis (called z-scores) was extended to physical and functional properties of proteins, where the proportion of particular amino acids rather than a precise sequence is the determining factor. Coomassie Brilliant Blue (CBB) spectral responses to amino acid homopolymers (R = 0.926) and proteins, either as a function of their contents of six basic and aromatic amino acids (R = 0.976) or as a function of the contributions of these amino acids to the three z-scores (R = 0.935), were modeled. The ultraviolet absorbance of proteins was modeled in terms of the z-score contributions of tyrosine, tryptophan and cysteine (R = 0.995). Modeling many protein functional properties in this manner appears possible. An approach to modeling peptide behaviors that depend on short sequences of amino acids was also considered.