The sequencing of the human genome and genomes from lower species has led molecular biology research into the era of proteomics. Starting from the cDNA’s, a large number of proteins have to be expressed in quantities large enough to allow determining of their structure and function, which forms the basis for the development of new pharmaceutical products. The larger number of these proteins is most probably new to us, and furthermore, we expect that they will involve a high-degree of complexity. The process development relies on ‘‘trial-and-error’’ methods to provide the bases for the choice of production method since we still do not have the knowledge that could guide this choice from the cDNA structure alone. We must therefore expect that the number and complexity of the proteins will require an equally wide portfolio of production methods. For this purpose, there is thus shown a large interest in new production systems, which could be used for a multitude of proteins as well as for a unique protein. Proteins for crystallisation and functional studies must be structurally correct and produced to a high degree of purity. In the valuation of recombinant systems for the production of proteins, the emphasis is therefore also placed on product quality. This issue of Bioprocess and Biosystems Engineering includes a summary of the research on a new production system based on cells from a moss. The system uses a strain, which has been genetically engineered to produce products with strictly human-specific glycosylation patterns (Decker and Reski), which gives great process advantages compared with the use of e.g., yeasts and animal cell systems. One might imagine that many years of research in bioprocess development would have led also to the development of general tools which could be used in the context of production for structural genomics too; but in many cases, these tools involve technologies which are not applicable on a small scale, since they are based on operator-supervised control systems of a technical complexity which cannot be used in a micro-scale, multi-parallel cultivation format. An example of such a very powerful control system is the fed-batch technique. This is the only cultivation technique, which allows the accumulation of cells to a high cell density. But in addition, this technique leads further to the establishment of different growth rates, which has been shown to be an important factor for the control of both productivity and product quality, particularly regarding the solubility of a protein. To achieve this on a small scale, a technology is proposed in this issue (Backlund et al.), where cell engineering is used to enable the control of substrate limitation at cell rather than at reactor level. There are great expectations that the large investments in time and money in the genome and proteome projects will lead to an increased number of new and better recombinant bioproducts. In those cases where these products are intended for use as biopharmaceuticals, great demands are made not only on the quality of the product but also on the reproducibility of the processes. This increases the demand for powerful analytical techniques capable of resolving quality steering issues very early on in the process as well as the appropriate control technology. Control algorithms are being developed in many lines of business outside the pharmaceutical sector, and the relatively weak development on the pharmaceutical G. Larsson (&) School of Biotechnology, Royal Institute of Technology, Roslagstullsbacken 21, Stockholm SE10691, Sweden e-mail: gen@biotech.kth.se