michael barton

@hirsh006
Our aim in this was to compare the relative effects of cost minimisation against CAI and tRNA count as predictors of expression level.

Our work showed that CAI is the major predictor of expression level, as much previous work has done. I agree that there would be less pressure to use abundant tRNAs for lowly expressed proteins, but the reverse would also be true for highly expressed proteins to require abundant tRNA. I think CAI is adaptation of highly expressed transcripts to use the optimal codons. So therefore the requirement for expression level would drive the evolution of codon adaptation.

Does this answer your question?

I'm confused by one thing. It seems like you are suggesting that the CAI causes the differences in expression level. Isn't it more likely that the reason for the correlation is that genes with lower expression levels have less pressure on them to use highly available tRNAs? Thus, expression levels cause the CAI.

It is interesting that there is such a strong effect of on free amino-acids but not on protein abundance. Still, all else equal there should be a slight pressure to reduce cost.