This paper discusses an improvement in a Stochastic Evolutionary Model of Protein Production Rate (SEMPPR) by revising the method by which it models mutation. SEMPPR previously assumed unbiased mutation, an assumption whose inaccuracy is made clear by observed codon counts of low-expression genes, where mutation determines equilibrium state. This paper presents a new, more complex model generalized on a per-codon basis and calculated from observed codon frequencies using a maximum likelihood framework. Results obtained from SEMPPR using the codon specific mutation model proved more accurate in predicting a protein’s production rate, reaffirming that complex mechanisms govern codon mutation rates.
"Improving Codon Evolution Models Using Complex Mutation Models,"
Pursuit - The Journal of Undergraduate Research at the University of Tennessee:
2, Article 5.
Available at: http://trace.tennessee.edu/pursuit/vol4/iss2/5