The fact is:
Eurofins’ optimisation software GENEius does not only use the best codons. We have shown that this very simple approach does not result in highest protein expression.
During the optimisation process GENEius randomly assembles the DNA sequence and then analyses it in relation to codon usage by comparing it to the input codon usage table. This input codon usage table is usually taken from the Kazusa Codon Usage Database (http://www.kazusa.or.jp/codon), but it can also be provided by the customer. GENEius does not simply aim for a high codon adaption index (CAI), instead it harmonises the codons used. Frequently used codons from highly expressed genes are more often used in the resulting gene sequence than less frequently used codons. Very rare codons, however, will be completely avoided. During adaption GENEius also checks for “bad motifs” like restriction sites and avoids artificial splice sites, unspecific transcription factor binding sites, etc. Also, to minimise RNA structure direct and inverted repeats are avoided as they not only make synthesis more difficult, they can decrease DNA stability and reduce efficiency of transcription and translation in E.coli. And last but not least, the GC content is equally distributed to avoid GC-rich subsequences within in the gene. All these parameters are taken into account and an “optimisation score” is constantly being calculated. If this score falls below a certain threshold, the sequence is taken as the output. This procedure results in a different DNA sequence every time the optimisation is running.