Metabolic engineering is the science of fine-tuning metabolic pathways using mathematical modelling. The aim is to simplify and quantify the complex chemical reactions within living systems, which are intertwined and occur as very precise fluxes of specific chemicals. When a recombinant organism doesn’t produce the product in the quantity or quality expected, it can be difficult to coax the metabolic pathway away from by-product creation. Maintaining the balance between the individual components involved in multiple, simultaneous reactions is a task that cells perform continuously -- so why is it so difficult for engineers to understand and manipulate?

SilicoLife, a spin-out from MIT Portugal program in bioengineering, is leveraging its computational biology platform to enable efficiencies residing at the intersection of computer sciences, life sciences, and engineering. The company hopes to carve out a unique niche within the growing list of industries tapping synthetic biology, including established companies that don’t have industrial biotechnology expertise and startups looking for a lean launch. Can SilicoLife prove the value of outsourcing such a critical component of research and development and become a key pillar to the field’s growing base of infrastructure? Or will the industry prefer to keep metabolic engineering capabilities in-house?

What SilicoLife does

Bioinformaticians and systems biologists use mathematical models to reduce large amounts of biological data into more digestible forms. While conventional chemical and biotech companies lack this approach, relying on random mutagenesis and using hit-and-miss trials; SilicoLife uses robust algorithms, mathematical modelling, and user-friendly software tools to optimize strains and bioprocesses.

Typical in-silico metabolic engineering workflow. Image source: SilicoLife

Genome sequence and literature, phenotypic, and physiological data are the prerequisite inputs to produce optimized strains for industrial purposes. Genome annotation is performed after genome sequencing and includes finding genes and working out their functions. Construction of genome-scale models is performed using a variety of techniques and involves retrieving and integrating data which is then complemented with data produced from validation experiments and that found in the literature. Metabolic engineering requires a special emphasis on localizing reactions and energy and mass balances at the molecular level. The model is then refined and validated to improve its predictive quality. Finally, the reconstructed and validated model is used to identify specific targets (such as genes to add or subtract) that maximize the productivity of the desired process.

SilicoLife has the expertise in dynamic annotation required for aspects relevant for metabolic modelling and engineering. They also specialize in databases and data integration to handle the biological data required. The capability to select the most appropriate optimization algorithm for each particular application allows the company to fine-tune metabolic pathways. Providing the analogy of a GPS, SilicoLife reduces a chaotic map into the most efficient route.

Who needs better bacteria?

Partners and customers involve the world’s leading chemical, materials, and synthetic biology companies. These are mostly established companies that don’t have the expertise required for industrial biotech. Last year SilicoLife entered into an agreement with INVISTA, a manufacturer of industrial intermediates, polymers, and fibers for use in clothing and everyday products. The pair is working together to optimize microbial strains for bioprocesses that produce various industrial chemicals.

How does a partnership work? The development of a chemical usually involves three steps. First, the most suitable organism is screened and modeled. Next, heterologous pathways are screened and evaluated for respective yield opportunities. Lastly, the researchers come up with robust metabolic engineering solutions that allow for overproduction of the target chemical with minimum by-products.

SilicoLife also offers custom software development for special projects, contract-research, and consulting to its customers. Just as metabolic pathways are related, the success of the company is intertwined with the success of its clients -- something an elegant workflow aims to enable.

However, there is serious competition -- sometimes from within the very companies SilicoLife looks to partner with. The aim is to save the time, costs, and labor involved in making better bacteria and production processes, but partners must feel comfortable outsourcing such a critical development step. While outsourcing has the advantage of allowing companies to focus on the design of their product and downstream processing (rather than the finer details associated with microbial optimization), some companies may simply prefer to develop in-house capabilities for strain optimization.

New infrastructure or outcast?

The use of in-silico tools, such as those at the heart of SilicoLife’s computational biology platform, accelerates research and shortens the time it takes to bring a product to the market. These tools increase the productivity, yield, and specificity of the manufacturing bioprocess and saves investment, raw material, and operational costs during the process -- advantages that could enable the rise of leaner biotech startups. Will the cost advantages be significant enough to catalyze the shift from biotech as a niche industry to a common manufacturing practice? We’re about to find out.