Today I learnt few concepts in protein expression which I found very interesting. As we all know, because of the redundancy of the triplet code, it is possible to preserve aminoacid sequence coding while varying the nucleic acid code. This can even happen as a silent mutation. However the availability of corresponding tRNA for each triplet varies from species to species. The translational efficiency of each code may be different (some times to a very great extend) though all may code for the same product. All silent mutations may not be silent after all.
Few codon pairs exist in relative abundance which may act as translation pause sites and slow down translational process. The tRNAs that bind during the translation of such a biased pair appear somehow incompatible. These codon pairs vary from species to species.
These points need to be considered while designing a gene for expression systems. Designing and implementing an algorithm to optimize the gene with respect to the above points can be a good bioinformatics project.