Genetic engineering constraints

The rules that have always governed the use of ethanol, government policy favoring one agricultural raw material over another, the new constraints that limit the marketing of genetically engineered products—all these factors serve to remind those interested in the... 

Other constraints are important for more complex products, for which mass is not central, but value. For example, central nervous system stimulants are a new class of substances addressed by production with engineered baker s yeast. Expressing the biosynthetic pathways for the opioids thebaine and hydrocodone, and parts of the morphine pathway in yeast, a first step is taken for easy production of opiates [21, 22]. This opens the possibility for the development of new painldllers with less addictive potential. However, it clearly is a new technology that could be abused with many negative consequences - so some contemplation about how to control these developments seems advisable [23]. Not only are narcotics in the center of interest but stimulants such as caffeine and theobromine have also recently been produced with genetically engineered 5. cerevisiae strains [24]. 

Reverse engineering has been demonstrated to work in principle for model genetic networks of binary genes connected through logical rules [18]. A key issue is the data requirement necessary to provide sufficient information to capture the complexity of the molecular network. In model networks it has been shown that only a tiny subset of all possible behaviors need to be known in order to infer network architecture with accuracy [18], provided that the network exhibits significant constraints (biomolecular networks are far removed from randomly connected networks) [20]. 

Genetic algorithms have become the method of choice in difficult engineering design circumstances where complex feasibility limitations and massive nonlinearity make it difficult to employ neighborhood-based methods. However, other methods usually give better performance on classic combinatorial optimization problems—especially ILP) s and cases with hnear constraints. 

---