The main goal of this study was to test the ability of an artificial neural network (ANN) for rice quality prediction based on grain physical parameters and to conduct a comparison with multiple linear regression (MLR) using 66 samples in duplicate. The parameters used for rice quality prediction are related to biochemical composition (starch, amylose, ash, fat, and protein concentration) and pasting parameters (peak viscosity, trough, breakdown, final viscosity, and setback). These parameters were estimated based on grain appearance (length, width, length/width ratio, total whiteness, vitreous whiteness, and chalkiness), and milling yield (husked, milled, head) data. The MLR models were characterized by very low coefficient determination (R² = 0.27–0.96) and a root-mean-square error (RMSE) (0.08–0.56). Meanwhile, the ANN models presented a range for R² = 0.97–0.99, being characterized for R² = 0.98 (training), R² = 0.88 (validation), and R² = 0.90 (testing). According to these results, the ANN algorithms could be used to obtain robust models to predict both biochemical and pasting profiles parameters in a fast and accurate form, which makes them suitable for application to simultaneous qualitative and quantitative analysis of rice quality. Moreover, the ANN prediction method represents a promising approach to estimate several targeted biochemical and viscosity parameters with a fast and clean approach that is interesting to industry and consumers, leading to better assessment of rice classification for authenticity purposes.

Biocatalysis is currently a workhorse used to produce a wide array of compounds, from bulk to fine chemicals, in a green and sustainable manner. The success of biocatalysis is largely thanks to an enlargement of the feasible chemical reaction toolbox. This materialized due to major advances in enzyme screening tools and methods, together with high-throughput laboratory techniques for biocatalyst optimization through enzyme engineering. Therefore, enzyme-related knowledge has significantly increased. To handle the large number of data now available, computational approaches have been gaining relevance in biocatalysis, among them machine learning methods (MLMs). MLMs use data and algorithms to learn and improve from experience automatically. This review intends to briefly highlight the contribution of biocatalysis within biochemical engineering and bioprocesses and to present the key aspects of MLMs currently used within the scope of biocatalysis and related fields, mostly with readers non-skilled in MLMs in mind. Accordingly, a brief overview and the basic concepts underlying MLMs are presented. This is complemented with the basic steps to build a machine learning model and followed by insights into the types of algorithms used to intelligently analyse data, identify patterns and develop realistic applications in biochemical engineering and bioprocesses. Notwithstanding, and given the scope of this review, some recent illustrative examples of MLMs in protein engineering, enzyme production, biocatalyst formulation and enzyme screening are provided, and future developments are suggested. Overall, it is envisaged that the present review will provide insights into MLMs and how these are major assets for more efficient biocatalysis.

The development of biopharmaceutical manufacturing processes presents critical constraints, with the major constraint being that living cells synthesize these molecules, presenting inherent behavior variability due to their high sensitivity to small fluctuations in the cultivation environment. To speed up the development process and to control this critical manufacturing step, it is relevant to develop high-throughput and in situ monitoring techniques, respectively. Here, high-throughput mid-infrared (MIR) spectral analysis of dehydrated cell pellets and in situ near-infrared (NIR) spectral analysis of the whole culture broth were compared to monitor plasmid production in recombinant Escherichia coli cultures. Good partial least squares (PLS) regression models were built, either based on MIR or NIR spectral data, yielding high coefficients of determination ( R²) and low predictive errors (root mean square error, or RMSE) to estimate host cell growth, plasmid production, carbon source consumption (glucose and glycerol), and by-product acetate production and consumption. The predictive errors for biomass, plasmid, glucose, glycerol, and acetate based on MIR data were 0.7 g/L, 9 mg/L, 0.3 g/L, 0.4 g/L, and 0.4 g/L, respectively, whereas for NIR data the predictive errors obtained were 0.4 g/L, 8 mg/L, 0.3 g/L, 0.2 g/L, and 0.4 g/L, respectively. The models obtained are robust as they are valid for cultivations conducted with different media compositions and with different cultivation strategies (batch and fed-batch). Besides being conducted in situ with a sterilized fiber optic probe, NIR spectroscopy allows building PLS models for estimating plasmid, glucose, and acetate that are as accurate as those obtained from the high-throughput MIR setup, and better models for estimating biomass and glycerol, yielding a decrease in 57 and 50% of the RMSE, respectively, compared to the MIR setup. However, MIR spectroscopy could be a valid alternative in the case of optimization protocols, due to possible space constraints or high costs associated with the use of multi-fiber optic probes for multi-bioreactors. In this case, MIR could be conducted in a high-throughput manner, analyzing hundreds of culture samples in a rapid and automatic mode.