Search

Aims: Precise implant positioning, tailored to individual spinopelvic biomechanics and phenotype, is paramount for stability in total hip arthroplasty (THA). Despite a few studies on instability prediction, there is a notable gap in research utilizing artificial intelligence (AI). The objective of our pilot study was to evaluate the feasibility of developing an AI algorithm tailored to individual spinopelvic mechanics and patient phenotype for predicting impingement. Methods: This international, multicentre prospective cohort study across two centres encompassed 157 adults undergoing primary robotic arm-assisted THA. Impingement during specific flexion and extension stances was identified using the virtual range of motion (ROM) tool of the robotic software. The primary AI model, the Light Gradient-Boosting Machine (LGBM), used tabular data to predict impingement presence, direction (flexion or extension), and type. A secondary model integrating tabular data with plain anteroposterior pelvis radiographs was evaluated to assess for any potential enhancement in prediction accuracy. Results: We identified nine predictors from an analysis of baseline spinopelvic characteristics and surgical planning parameters. Using fivefold cross-validation, the LGBM achieved 70.2% impingement prediction accuracy. With impingement data, the LGBM estimated direction with 85% accuracy, while the support vector machine (SVM) determined impingement type with 72.9% accuracy. After integrating imaging data with a multilayer perceptron (tabular) and a convolutional neural network (radiograph), the LGBM’s prediction was 68.1%. Both combined and LGBM-only had similar impingement direction prediction rates (around 84.5%). Conclusion: This study is a pioneering effort in leveraging AI for impingement prediction in THA, utilizing a comprehensive, real-world clinical dataset. Our machine-learning algorithm demonstrated promising accuracy in predicting impingement, its type, and direction. While the addition of imaging data to our deep-learning algorithm did not boost accuracy, the potential for refined annotations, such as landmark markings, offers avenues for future enhancement. Prior to clinical integration, external validation and larger-scale testing of this algorithm are essential. Cite this article: Bone Jt Open 2024;5(8):671–680.

Abstract Background Amyrin is an important triterpenoid and precursor to a wide range of cosmetic, pharmaceutical and nutraceutical products. In this study, we metabolically engineered the oleaginous yeast, Yarrowia lipolytica to produce α- and β-amyrin on simple sugar and waste cooking oil. Results We first validated the in vivo enzymatic activity of a multi-functional amyrin synthase (CrMAS) from Catharanthus roseus, by expressing its codon-optimized gene in Y. lipolytica and assayed for amyrins. To increase yield, prevailing genes in the mevalonate pathway, namely HMG1, ERG20, ERG9 and ERG1, were overexpressed singly and in combination to direct flux towards amyrin biosynthesis. By means of a semi-rational protein engineering approach, we augmented the catalytic activity of CrMAS and attained ~ 10-folds higher production level on glucose. When applied together, protein engineering with enhanced precursor supplies resulted in more than 20-folds increase in total amyrins. We also investigated the effects of different fermentation conditions in flask cultures, including temperature, volumetric oxygen mass transfer coefficient and carbon source types. The optimized fermentation condition attained titers of at least 100 mg/L α-amyrin and 20 mg/L β-amyrin. Conclusions The design workflow demonstrated herein is simple and remarkably effective in amplifying triterpenoid biosynthesis in the yeast Y. lipolytica.

Abstract Background In biological cells, promoters drive gene expression by specific binding of RNA polymerase. They determine the starting position, timing and level of gene expression. Therefore, rational fine-tuning of promoters to regulate the expression levels of target genes for optimizing biosynthetic pathways in metabolic engineering has recently become an active area of research. Results In this study, we systematically detected and characterized the common promoter elements in the unconventional yeast Yarrowia lipolytica, and constructed an artificial hybrid promoter library that covers a wide range of promoter strength. The results indicate that the hybrid promoter strength can be fine-tuned by promoter elements, namely, upstream activation sequences (UAS), TATA box and core promoter. Notably, the UASs of Saccharomyces cerevisiae promoters were reported for the first time to be functionally transferred to Y. lipolytica. Subsequently, using the production of a versatile platform chemical isoamyl alcohol as a test study, the hybrid promoter library was applied to optimize the biosynthesis pathway expression in Y. lipolytica. By expressing the key pathway gene, ScARO10, with the promoter library, 1.1–30.3 folds increase in the isoamyl alcohol titer over that of the control strain Y. lipolytica Po1g KU70∆ was achieved. Interestingly, the highest titer increase was attained with a weak promoter P UAS1B4-EXPm to express ScARO10. These results suggest that our hybrid promoter library can be a powerful toolkit for identifying optimum promoters for expressing metabolic pathways in Y. lipolytica. Conclusion We envision that this promoter engineering strategy and the rationally engineered promoters constructed in this study could also be extended to other non-model fungi for strain improvement.

Itaconic acid (IA) is a high-value organic acid with a plethora of industrial applications. In this study, we seek to develop a microbial cell factory that could utilize waste cooking oil (WCO) as raw material for circular and cost-effective production of the abovementioned biochemical. Specifically, we expressed cis-aconitic acid decarboxylase (CAD) gene from Aspergillus terreus in either the cytosol or peroxisome of Yarrowia lipolytica and assayed for production of IA on WCO. To further improve production yield, the 10 genes involved in the production pathway of acetyl-CoA, an intermediate metabolite necessary for the synthesis of cis-aconitic acid, were individually overexpressed and investigated for their impact on IA production. To minimize off-target flux channeling, we had also knocked out genes related to competing pathways in the peroxisome. Impressively, IA titer up to 54.55 g/L was achieved in our engineered Y. lipolytica in a 5 L bioreactor using WCO as the sole carbon source.

With the continuous expansion of power grid, the system structure becomes more and more complex, and multiple faults occur frequently. And multiple faults are the key and difficulty of fault diagnosis. Due to the huge and complex power grid structure and the large data size of fault processing, it is necessary to diagnose the power grid fault quickly and accurately. In this paper, based on convolutional neural network, a multi-fault diagnosis model of distribution network based on fuzzy optimal convolutional neural network is proposed. Firstly, fault line and fault type judgment based on two soft maximum classifiers are analyzed. Membership functions of distribution network faults are established by using fuzzy theory. Secondly, the influence of convolution kernel number and sample width on the accuracy of model diagnosis is studied and analyzed. Simulation results show that, under the same conditions, the accuracy of fuzzy optimized convolutional neural network for multiple fault diagnosis is higher than that of convolutional neural network. The time of fault diagnosis and training is less than that of convolutional neural network. [ABSTRACT FROM AUTHOR]