Biomanufacturing

Biomanufacturing is the production of commercially valuable molecules using engineered biological systems — typically microorganisms like Escherichia coli or Saccharomyces cerevisiae, but increasingly mammalian cells, plant cells, and cell-free systems. By reprogramming cellular metabolism through synthetic biology and metabolic engineering, biomanufacturing enables the sustainable production of compounds that are difficult, expensive, or environmentally harmful to synthesize chemically ¹.

How It Works

Biomanufacturing follows a design-build-test-learn (DBTL) cycle:

1. Pathway design: Identify or engineer a metabolic route from a cheap feedstock (glucose, glycerol, CO2) to the target molecule. This may involve importing heterologous enzymes from other organisms, knocking out competing pathways, or rewiring regulatory networks
2. Strain engineering: Introduce the designed genetic modifications into a production host using tools like CRISPR-Cas9, homologous recombination, or plasmid-based expression systems
3. Screening and optimization: High-throughput screening (microplate assays, biosensors, FACS-based selection) identifies top-performing strain variants from combinatorial libraries
4. Scale-up: Transfer optimized strains from shake flask or microbioreactor conditions to pilot-scale (10-100 L) and production-scale (10,000-200,000 L) fermenters

Computational Considerations

Modern biomanufacturing is computationally intensive at every stage:

• Flux balance analysis (FBA): Constraint-based models of genome-scale metabolic networks (GEMs) predict theoretical yields, identify gene knockout targets, and evaluate pathway feasibility before any wet-lab work ²
• Retrosynthesis algorithms: Tools like RetroPath and MINE enumerate possible enzymatic routes from target molecule back to available precursors, scoring pathways by thermodynamic feasibility and enzyme availability
• Machine learning for strain optimization: Models trained on combinatorial expression data (promoter strength, RBS variants, copy number) predict optimal genetic configurations, reducing the number of build-test cycles required
• Process digital twins: Physics-informed models of bioreactor dynamics (oxygen transfer, pH, temperature, substrate feeding) enable real-time optimization of fermentation conditions and predictive scale-up

Landmark Examples

Several biomanufacturing successes demonstrate the field’s impact:

• Artemisinin (antimalarial): Keasling’s group engineered S. cerevisiae to produce artemisinic acid, the precursor to the antimalarial drug artemisinin, from glucose — reducing dependence on agricultural supply chains ³
• 1,3-Propanediol (materials): DuPont and Genencor engineered E. coli to produce 1,3-PDO from corn sugar for Sorona polymer fiber, one of the first large-scale industrial bioprocesses
• Insulin (therapeutics): Recombinant human insulin produced in E. coli (Humulin, approved 1982) was the first commercially available product of genetic engineering
• Impossible Burger (food): Soy leghemoglobin produced by engineered Pichia pastoris gives the Impossible Burger its meat-like flavor and color

Scale and Economics

The economics of biomanufacturing depend on several factors:

• Titer: The concentration of product in the fermentation broth (g/L). Commercial viability for bulk chemicals typically requires titers above 50-100 g/L
• Rate: Volumetric productivity (g/L/h) determines capital utilization and throughput
• Yield: The fraction of feedstock carbon converted to product (g product/g substrate). Theoretical maximum yields are constrained by stoichiometry and thermodynamics
• Downstream processing: Separation and purification of the product from the fermentation broth often accounts for 50-80% of total production cost

Limitations

• Metabolic burden: Diverting cellular resources toward product synthesis competes with growth, creating an evolutionary pressure to lose production capacity over long fermentations
• Scale-up challenges: Heterogeneous mixing, oxygen gradients, and shear stress in large bioreactors create microenvironments absent in laboratory conditions
• Pathway complexity: Multi-step pathways involving 10+ heterologous enzymes require extensive balancing of expression levels to avoid bottlenecks and toxic intermediate accumulation
• Regulatory timelines: Biomanufactured products for food, pharmaceutical, and agricultural markets face lengthy regulatory approval processes

---

Woolf Software builds computational models for metabolic pathway design, strain optimization, and fermentation process modeling. Get in touch.