Variance component analysis of polymorphic metabolic systems.

The relationship between mechanistic allelic interaction in multi-gene systems and genetic contribution to population variance remains poorly understood. In order to address this problem, dynamic cellular processes must be reconciled with individual differences in a population. We suggest an approach to enable this for metabolic systems, whereby steady-state biomarker concentrations are calculated for individual systems carrying different alleles. As proof of principle, we simulated two versions of a three enzyme linear synthesis pathway, in a multi-level framework from transcription to enzyme action. The first (Standard) model incorporated conventional kinetics, whilst an analogous model included negative feedback in the form of competitive inhibition (CI). Alleles were allowed to confer different transcription rates, and genetic components of variance in biomarker concentration calculated for populations of each system type. Initial simulations of high and low expression alleles revealed substantial genetic additivity and some dominance for both system architectures. For the Standard model population, each of the three genes contributed equally, whereas CI substantially altered the relative importance of individual genes. Epistasis was limited for both model populations, never rising above 5% in extensive parameter explorations. Subsequent simulations examined a wide range of allelic transcription rates, from almost null to three orders of magnitude above baseline. Again, for both model architectures, additive and dominance effects were most prevalent, but epistasis increased substantially as allelic effects approached null. We conclude that the nature of allelic contribution to variance is dependent upon both the magnitude of conferred effect and the structure of the system in which it is embedded, and relatively little on other system parameters. We believe that implementation of this approach holds promise of a better understanding of the genotype to phenotype transition, particularly in integrating small allelic effects into larger frameworks.