Biotech Values

[genisi]

The phenomenon is intriguing, but I do wonder whether the correlation is over-interpreted given the complexity of human genomes

While it is already known that synonymous changes in codons are not so silent as once thought, you are rightly wondering about the importance of the phenomenon. Not enough work has been done yet but this one by Chen et. al. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2962641/?tool=pubmed found that nsSNPs and sSNPs shared similar likelihood and effect size for disease association.
One note on the the measurement of relative synonymous codon usage (RSCU) used in the paper before I go on:
RSCU values show the number of times a particular codon is observed relative to the number of times that the codon would be observed if all the codons for a given amino acid had the same probability.
Change in RSCU = deltaRSCU, might be associated with a change in local translation elongation rates i.e. the speed of translation. So it represents only this parameter and does not capture others like the location of the mutation, failure of splicing, exon skipping, mRNA stability, microRNA binding, all of which can also be crucial.
My view is that the surface has only been scratched but I think synonymous changes in codons can sometimes have a dramatic effect on phenotypic outcome.