The thermodynamic stability hypothesis DNA results

Several lines of evidence support the thermodynamic stability hypothesis at the DNA level. While their strength varies from weak to strong, their cumulative weight in favor of the hypothesis is overwhelming. 

In contradiction with this conclusion, a plot of the maximal (or mean) body tempera-liii es of cold-blooded vertebrates against the standard deviation of their GC3 values (data of Belle ct al., 2003) showed a significant (yet. admittedly, borderline) correlation, provided that the data from alligator and turtle were neglected. This elimination is justified by the fact that these species were not only represented by a very low number of genes (16 and 17), but also showed the largest and smallest standard deviations, respectively (see Fig, 12.7), in 

A conclusion suggested by the results of Fig. 12.7 is that the compositional heterogeneity of third codon positions of vertebrates (which essentially depends upon the relative amounts of GC-rich genes) may well be correlated with body temperature. While this approach can, and will, be pursued further, comparisons of aligned orthologous genes are needed to provide a final proof for the differences. 

A suggestion derived from the above results and considerations is that the compositional heterogeneity of both DNA and third codon positions could be compared with CpG and methylation levels instead of being compared with body temperature. Indeed, non only can CpG and methylation be precisely measured, but these parameters are likely to integrate time and temperature effects, thus solving the major problem encountered when using only direct temperature measurement in this kind of experimental approach. 

Figui e 12.8. CpG levels ace plotted against GC levels for 40 kb genome segments of Drosophila, zebrafish. pulfci fisl), platypus, and human. (From Jabbari and Bernardi. 2004b). 

---