In order to optimize strains for higher production yields of desired products or compounds, computational tools have been developed and employed to guide the design of knockout-strains with modified metabolic pathways leading to an increase in the bioproduct production. Two major examples of such tools are OptKnock (Burgard et al. 2003) and OptStrain (Pharkya et al. 2004). The OptKnock tool, for example, was used to identify gene knockout strategies leading to the optimization of bacterial strains for higher yields of lactate, succinate, and 1, 3-propanediol productions. In a similar approach, the Optknock tool has also allowed the

optimization of E. coli strains to achieve a higher production of several amino acids and has set in place a set of knockout strategies to achieve this goal (Pharkya et al. 2003). Pharkya et al. have established several different gene knockout combinations with each resulting in an increased production of a specific amino acid. For example, a knockout of three genes coding for three enzymes, namely pyruvate dehydrogenase, pyruvate formate lyase, and an ATPase, has allowed the cells to achieve a production of 14.95 mmol/g DW h of alanine. Alternatively, another strategy involving the knockout of a fourth enzyme, phosphofructokinase, on top of the previously mentioned genes, has increased the alanine production to 18.53 mmol/g DW h. Although the actual amino acid production has increased, the growth rate of the organism has sharply decreased and the choice of which strategy to employ remains subject to the investigator’s preferences and judgment, and is limited by the experimental setup limitations.

Lastly, OptStrain is a tool that can help identify the optimal strains producing the desired metabolite by determining the best substrate choices and genes that need to be deleted or over-expressed in order to achieve the increase in metabolite pro­duction. This strategy ensures high growth rates while optimizing the production of the desired bio-product.

A different approach, metabolic transformation algorithm (or MTA) (Yizhak et al. 2013), can potentially be used to shift the phenotypic state of algae from, e. g., growth with low lipid production, to a state at which lipids are produced at high rates with or without nutritional stress. Yizhak and colleagues introduced this algorithm that was used to predict gene knockouts that can shift metabolism from a given “source” state to a desired “target” state. The approach uses gene expression profiles of the two states in predicting gene deletions that forces changes in the flux distribution of source state to match the desired target state. While the authors used the yeast system to validate their algorithm, the approach is not system-, or state — specific, and is applicable to any organism, and many different phenotypic states. This algorithm presents a potentially exciting method for altering algal metabolism from a nonproductive to a productive state.