Boosting corn stover bioconversion to fatty acids with Gaussian Process optimization

The conversion of lignocellulosic biomass into valuable products such as fatty acids is promising for sustainable bioeconomy development. However, challenges in optimizing bioprocess parameters and addressing inhibitory by-products have limited the economic feasibility. This study employed Gaussian Process (GP) optimization to enhance gamma-linolenic acid (GLA) production from corn stover and docosahexaenoic acid (DHA) fermentation. Following an evaluation of three pretreatment methods—dilute alkali, deep eutectic solvent, and dilute acid—GP optimization refined enzymatic hydrolysis and fermentation parameters, outperforming traditional response surface methodology (RSM). Detoxification strategies using activated carbon, combined with GP guidance, further improved lipid yields. Techno-economic analysis (TEA) showed a competitive minimum selling price of $17,821 per ton for GLA oil, with sensitivity analysis revealed key cost drivers, and suggesting future avenues for cost reduction. This highlights the benefits of integrating machine learning (ML) with bioprocess engineering, providing a scalable approach to optimize biomass conversion for industrial applications. The primary challenge in the industrial application of lignocellulose is to optimize the process parameters and remove the inhibitors. Combining experimental research and process simulation, herein, we developed an machine learning-assisted corn stover conversion process aimed at improving lipid yield and maximizing techno-economic feasibility. • Employing Gaussian Process optimization to enhance gamma-linolenic acid production from corn stover and docosahexaenoic acid fermentation. • Detoxification strategies using activated carbon, combined with GP guidance, further improved lipid yields. • GP optimization boosts polyunsaturated fatty acids production by 17 %, compared to commonly used response surface methodology. • Techno-economic analysis showed a competitive selling price of GLA oil and revealed three key cost drivers.