Opportunities and challenges of machine learning in bioprocesses: Categorization from different perspectives and future direction

Explainable machine learning for arsenic remobilization potential in the vadose zone: Leveraging readily available soil properties

The vadose zone acts as a natural buffer that prevents contaminants such as arsenic (As) from contaminating groundwater resources. Despite its capability to retain As, our previous studies revealed that a substantial amount of As could be remobilized from soil under repeated wet-dry conditions. Overlooking this might underestimate the potential risk of groundwater contamination. This study quantified the remobilization of As in the vadose zone and developed a prediction model based on soil properties. 22 unsaturated soil columns were used to simulate vadose zones with varying soil properties. Repeated wet-dry cycles were conducted upon the As-retaining soil columns. Consequently, 13.9-150.6 mg/kg of As was remobilized from the columns, which corresponds to 37.0-74.6 % of initially retained As. From the experimental results, a machine learning model using a random forest algorithm was established to predict the potential for As remobilization based on readily accessible soil properties, including organic matter (OM) content, iron (Fe) content, uniformity coefficient, D30, and bulk density. Shapley additive explanation analyses revealed the interrelated effects of multiple soil properties. D30, which is inter-related with Fe content, exhibited the highest contribution to As remobilization, followed by OM content, which was partially mediated by bulk density.

Nonionic surfactant Tween 80-facilitated bacterial transport in porous media: A nonmonotonic concentration-dependent performance, mechanism, and machine learning prediction

The surfactant-enhanced bioremediation (SEBR) of organic-contaminated soil is a promising soil remediation technology, in which surfactants not only mobilize pollutants, but also alter the mobility of bacteria. However, the bacterial response and underlying mechanisms remain unclear. In this study, the effects and mechanisms of action of a selected nonionic surfactant (Tween 80) on Pseudomonas aeruginosa transport in soil and quartz sand were investigated. The results showed that bacterial migration in both quartz sand and soil was significantly enhanced with increasing Tween 80 concentration, and the greatest migration occurred at a critical micelle concentration (CMC) of 4 for quartz sand and 30 for soil, with increases of 185.2% and 27.3%, respectively. The experimental results and theoretical analysis indicated that Tween 80-facilitated bacterial migration could be mainly attributed to competition for soil/sand surface sorption sites between Tween 80 and bacteria. The prior sorption of Tween 80 onto sand/soil could diminish the available sorption sites for P. aeruginosa, resulting in significant decreases in deposition parameters (70.8% and 33.3% decrease in KD in sand and soil systems, respectively), thereby increasing bacterial transport. In the bacterial post-sorption scenario, the subsequent injection of Tween 80 washed out 69.8% of the bacteria retained in the quartz sand owing to the competition of Tween 80 with pre-sorbed bacteria, as compared with almost no bacteria being eluted by NaCl solution. Several machine learning models have been employed to predict Tween 80-faciliated bacterial transport. The results showed that back-propagation neural network (BPNN)-based machine learning could predict the transport of P. aeruginosa through quartz sand with Tween 80 in-sample (2 CMC) and out-of-sample (10 CMC) with errors of 0.79% and 3.77%, respectively. This study sheds light on the full understanding of SEBR from the viewpoint of degrader facilitation.

Predicting maturity and identifying key factors in organic waste composting using machine learning models

The measurement of germination index (GI) in composting is a time-consuming and laborious process. This study employed four machine learning (ML) models, namely Random Forest (RF), Artificial Neural Network (ANN), Support Vector Regression (SVR), and Decision Tree (DT), to predict GI based on key composting parameters. The prediction results showed that the coefficient of determination (R2) for RF (>0.9) and ANN (>0.9) was higher than SVR (<0.6) and DT (<0.8), suggesting that RF and ANN displayed superior predictive performance for GI. The SHapley additive exPlanations value result indicated that composting time, temperature, and pH were the important features contributing to GI. Composting time was found to have the most significant impact on GI. Overall, RF and ANN were suggested as effective tools for predicting GI in composting. This study offers the reliable approach of accurately predicting GI in composting processes, thereby enabling intelligent composting practices.

Effects of Bacillus-based inoculum on odor emissions co-regulation, nutrient element transformations and microbial community tropological structures during chicken manure and sawdust composting

This study aims to evaluate whether different doses of Bacillus-based inoculum inoculated in chicken manure and sawdust composting will provide distinct effects on the co-regulation of ammonia (NH3) and hydrogen sulfide (H2S), nutrient conversions and microbial topological structures. Results indicate that the Bacillus-based inoculum inhibits NH3 emissions mainly by regulating bacterial communities, while promotes H2S emissions by regulating both bacterial and fungal communities. The inoculum only has a little effect on total organic carbon (TOC) and inhibits total sulfur (TS) and total phosphorus (TP) accumulations. Low dose inoculation inhibits total potassium (TK) accumulation, while high dose inoculation promotes TK accumulation and the opposite is true for total nitrogen (TN). The inoculation slightly affects the bacterial compositions, significantly alters the fungal compositions and increases the microbial cooperation, thus influencing the compost substances transformations. The microbial communities promote ammonium nitrogen (NH4+-N), TN, available phosphorus (AP), total potassium (TK) and TS, but inhibit nitrate nitrogen (NO3--N), TP and TK. Additionally, the bacterial communities promote, while the fungal communities inhibit the nitrite nitrogen (NO2--N) production. The core bacterial and fungal genera regulate NH3 and H2S emissions through the secretions of metabolic enzymes and the promoting or inhibiting effects on NH3 and H2S emissions are always opposite. Hence, Bacillus-based inoculum cannot regulate the NH3 and H2S emissions simultaneously.

An artificial intelligence approach for identification of microalgae cultures

In this work, a model for the characterization of microalgae cultures based on artificial neural networks has been developed. The characterization of microalgae cultures is essential to guarantee the quality of the biomass, and the objective of this work is to achieve a simple and fast method to address this issue. Data acquisition was performed using FlowCam, a device capable of capturing images of the cells detected in a culture sample, which are used as inputs by the model. The model can distinguish between 6 different genera of microalgae, having been trained with several species of each genus. It was further complemented with a classification threshold to discard unwanted objects while improving the overall accuracy of the model. The model achieved an accuracy of up to 97.27% when classifying a culture. The results demonstrate the effectiveness of the Deep Learning models for the characterization of microalgae cultures, it being a useful tool for the monitoring of microalgae cultures in large-scale production facilities while providing accurate characterization over a wide range of genera.

Artificial intelligence and machine learning for smart bioprocesses

In recent years, the digital transformation of bioprocesses, which focuses on interconnectivity, online monitoring, process automation, artificial intelligence (AI) and machine learning (ML), and real-time data acquisition, has gained considerable attention. AI can systematically analyze and forecast high-dimensional data obtained from the operating dynamics of bioprocess, allowing for precise control and synchronization of the process to improve performance and efficiency. Data-driven bioprocessing is a promising technology for tackling emerging challenges in bioprocesses, such as resource availability, parameter dimensionality, nonlinearity, risk mitigation, and complex metabolisms. This special issue entitled "Machine Learning for Smart Bioprocesses (MLSB-2022)" was conceptualized to incorporate some of the recent advances in applications of emerging tools such as ML and AI in bioprocesses. This VSI: MLSB-2022 contains 23 manuscripts, and summarizes the major findings that can serve as a valuable resource for researchers to learn major advances in applications of ML and AI in bioprocesses.