Developing high-dimensional machine learning models to improve generalization ability and overcome data insufficiency for mixed sugar fermentation simulation

Advances in the biological production of sugar alcohols from biomass-derived xylose

Sugar alcohols are a common class of low-calorie sweeteners. The advancement of technologies utilizing renewable resources has heightened interest in synthesizing sugar alcohols from biomass-derived xylose for cost down of process and sustainability. This review focuses on the potential of biomass-derived xylose and its effective conversion into sugar alcohols, underscoring the significance of this process in sustainable industrial applications. The two main approaches for producing sugar alcohols which include enzyme catalysis and microbial fermentation are thoroughly discussed. The microbial fermentation pathway relies on genetically engineered strains, which are modified to efficiently convert xylose into target sugar alcohols. Enzyme catalysis, on the other hand, directly converts xylose to sugar alcohols through specific reactions. In addition, strategies to improve product selectivity and reduce by-products are discussed in the paper, which are crucial for improving the economic viability and environmental sustainability of sugar alcohol production. Overall, utilizing xylose from biomass to produce sugar alcohols manifests environmental and economic benefits, indicating its substantial potential in the shift towards a low-carbon economy. Future studies may further explore cutting edge technologies to maximize the utilization of biomass-derived xylose and the sustainable production of sugar alcohols.

A comprehensive review on the application of neural network model in microbial fermentation

The development of high-performance strains and the continuous breakthrough of strain screening technology also pose challenges to downstream fermentation optimization and scale-up. Therefore, neural network models are utilized to optimize the fermentation process to meet the goals of boosting yield or lowering cost, with the use of artificial intelligence technology in conjunction with the peculiarities of the fermentation process. High-performance strains' yield rise and fermentation process amplification will be sped up with the aid of neural network models. This paper offers a helpful review for anyone interested in state-of-the-art microbial fermentation processes, as it thoroughly reviews the application of neural network models in predicting fermentation yield, optimizing the fermentation process, and monitoring the fermentation process.

Research advances on the consolidated bioprocessing of lignocellulosic biomass

Lignocellulosic biomass is an abundant and renewable bioresource for the production of biofuels and biochemical products. The classical biorefinery process for lignocellulosic degradation and conversion comprises three stages, i.e., pretreatment, enzymatic saccharification, and fermentation. However, the complicated pretreatment process, high cost of cellulase production, and insufficient production performance of fermentation strains have restricted the industrialization of biorefinery. Consolidated bioprocessing (CBP) technology combines the process of enzyme production, enzymatic saccharification, and fermentation in a single bioreactor using a specific microorganism or a consortium of microbes and represents another approach worth exploring for the production of chemicals from lignocellulosic biomass. The present review summarizes the progress made in research of CBP technology for lignocellulosic biomass conversion. In this review, different CBP strategies in lignocellulose biorefinery are reviewed, including CBP with natural lignocellulose-degrading microorganisms as the chassis, CBP with biosynthetic microorganisms as the chassis, and CBP with microbial co-culturing systems. This review provides new perspectives and insights on the utilization of low-cost feedstock lignocellulosic biomass for production of biochemicals.

The Efficacy of Machine Learning Models for Predicting the Prognosis of Heart Failure: A Systematic Review and Meta-Analysis

INTRODUCTION

Heart failure (HF) is a major global public health concern. The application of machine learning (ML) to identify individuals at high risk and enable early intervention is a promising approach for improving HF prognosis. We aim to systematically evaluate the performance and value of ML models for predicting HF prognosis.

METHODS

PubMed, Web of Science, Scopus, and Embase online databases were searched up to April 30, 2023, to identify studies on the use of ML models to predict HF prognosis. HF prognosis primarily encompasses readmission and mortality. The meta-analysis was conducted by MedCalc software. Subgroup analyses include grouping based on types of ML models, time intervals, sample sizes, the number of predictive variables, validation methods, whether to conduct hyperparameter optimization and calibration, data set partitioning methods.

RESULTS

A total of 31 studies were included. The most common ML models were random forest, boosting, support vector machine, neural network. The area under the receiver operating characteristic curve (AUC) for predicting HF readmission was 0.675 (95% CI: 0.651-0.699, p < 0.001), and the AUC for predicting HF mortality was 0.790 (95% CI: 0.765-0.816, p < 0.001). Subgroup analyses revealed that models with the prediction time interval of 1 year, sample sizes ≥10,000, the number of predictive variables ≥100, external validation, hyperparameter tuning, calibration adjustment, and data set partitioning using 10-fold cross-validation exhibited favorable performance within their respective subgroups.

CONCLUSION

The performance of ML models in predicting HF readmission is relatively poor, while its performance in predicting HF mortality is moderate. The quality of the relevant studies is generally low, it is essential to enhance the predictive capabilities of ML models through targeted improvements in practical applications.

Integration of genetic engineering and multi-factor fermentation optimization for co-production of carotenoid and DHA in Schizochytrium sp

Schizochytrium sp., a microalga with high lipid content, holds the potential for co-producing docosahexaenoic acid (DHA) and carotenoids. In this study, the ability of Schizochytrium sp. to naturally produce carotenoids was systematically explored. Further, by enhancing the precursor supply of geranylgeranyl diphosphate, regulating carbon source through sugar limitation fermentation and employing a combination of response surface methodology and artificial neural networks to precisely optimize nitrogen sources, a new record of 43-fold increase in β-carotene titer was achieved in the 5L bioreactor (653.2 mg/L). Meanwhile, a high DHA content was maintained (13.4 g/L). Furthermore, the use of corn stover hydrolysate has effectively lowered the production costs of carotenoid and DHA while sustaining elevated production levels (with total carotenoid titer and DHA titer reached 502.0 mg/L and 13.2 g/L, respectively). This study offers an efficient and cost-effective method for the co-production of carotenoid and DHA in Schizochytrium sp..