Depends on the type of analysis you want to do. You have to compare the functional properties of your recombinant construct to those of its natural form, e.g enzymatic activity. If large tags are used, frequently a protease site is introduced between tag and linker, allowing to cleave off the tag after purification. You will have to decide on the appropriate control experiments for the specific functions you are studying to show that your expression construct is an appropriate substitute for its native equivalent.

Dear Dr. Honegger,

Thanks for your answer and participation and comments. The question was about optimizing gene structure, mRNAs and proteins based on bioinformatics tools and after based on empirical essays and approaches by using linkers and few long tags such as hydrophobin and Elastin like peptides. Please look at this paper: An in silico chimeric multi subunit vaccine targeting virulence factors of enterotoxigenic Escherichia coli (ETEC) with its bacterial inbuilt adjuvant http://www.sciencedirect.com/science/article/pii/S0167701212001339

In fact, I am cooperating with a bioinformatician and we are progressing well on the project and hopefully we will submit in the next two weeks a Mini Review in a new journal called Plant science today about this topic. From other hand and at experimental level, there are many others criteria and factors such as host plant and expression cassette and chromatin structure.

My background is in design and application of gene libraries for in-vitro selection and evolution of binders, mainly prokaryotic expression systems, no specific experience in plant expression systems.

But based on my experience, the following are critical features:

- Vectors: viral, insertion in host genome ...

- Transcriptional regulation: inducible or constitutive, desired expression level

=> affecting how much RNA is produced

- Initiation of translation: At least in prokaryotic systems, the N-terminal sequence of the protein to be expressed can have a mayor influence on the efficiency on the initiation of translation, probably through RNA secondary structure. N-terminal tags help a lot to reduce these difference. In in-vivo systems, RNA sequence features affecting RNA stability may also have a major influence, here RNA secondary structure downstreams of the ORF may have a major influence

=> affecting how much protein is synthesized

- Targeting protein to a specific compartmens: cytoplasmic expression or signal sequences for export, accumulation in the vacuole

- Protein folding and stability, solubility, susceptibility to proteases, post-translational modifications either needed or to be avoided (disulfide bonds, glycosylations, phosphorylations, proteolytic processing. Fusions of folded domains may help keep the protein in solution, but do not really affect the stability of the desired domains. On the other hand, the more efficient chaperone systems of eukaryotic expression systems often reduces the differences seen in prokaryotic systems.

- Ease of production and of downstream processing, protein activity.

The quality and amount of protein produced depends on all of these steps - In experimentally optimizing production, which of the factors is limiting often depends on the particular protein construct.

Dear Dr. Annemarie Honegger,

I will for sure take into account your suggestions and please check the in silico article I published on /october 4 entitled 'An in silico overview on the usefulness of tags and linkers in plant molecular pharming".

I will send your suggestions to the bioinformatics specialist to think about new measurements to optimize and improve the whole expression cassette structure and function