Reconstructing editable CAD models from mesh data remains a fundamental yet challenging problem in reverse engineering. Existing methods often struggle to achieve both geometric accuracy and topological consistency, especially when recovering analytic design intent from complex meshes. We present an interactive CAD reverse modeling framework based on variational curve approximation, which bridges mesh segmentation and parametric reconstruction in a unified manner. Our method extracts surface-aligned cutting curves from segmented regions and refines them through a variational approximation process that encodes primitive attributes and geometric constraints. This enables the automatic recovery of closed loops, constraint relations, and operation parameters, leading to a compact and topologically consistent CAD representation. In addition, a coplanar profile detection and voxel-similarity-based Boolean inference are developed to restore the topological order of the original modeling sequence. Extensive experiments demonstrate that our approach achieves higher geometric fidelity, better structural consistency, and improved automation compared with state-of-the-art techniques. The reconstructed models are watertight, topologically consistent, and readily applicable to downstream CAD design tasks.

Copper-containing nitrite reductases (CuNiRs) catalyse the reduction of nitrite to nitric oxide and are a key enzyme in the anaerobic ammonium oxidation and denitrification steps of the nitrogen cycle. The recent recognition of the widespread distribution of three-domain CuNiRs where cognate redox partners are fused to the core NiR enzyme offered the possibility of studying coordinated events (e.g. proton-coupled electron transfer) in a conformationally stable donor-acceptor complex. The C-terminal cytochrome c tethered domain of the CuNiR from Ralstonia pickettii (RpNiR) has been well studied. Reverse engineering of RpNiR undertaken to remove the cognate partner domain showed that the presence of the additional domain resulted in significant differences in the apparent Km for nitrite and the reduction potentials of the Cu centres when compared with the core enzyme. The oxidation state of the haem centre and the position of the tethering linker have also been shown to control access of substrate to the active site. A key feature of this control is a conserved tyrosine residue (Tyr323 in RpNiR) located in the tethering linker between the fused domain and the core enzyme. To gain insight into this control, we have undertaken targeted mutations of RpNiR to probe the so-called primary proton channel and perturb putative electron transfer routes from the haem to the `gatekeeper' Tyr323 and to the T1Cu centre. The resolution of our crystallographic data to better than 1.2 Å enabled us to apply unrestrained SHELXL refinement of the structures. Our data provide a significant advance in our understanding of catalysis and modulation of electron transfer in these tethered systems, with wider implications for these fundamental processes in other protein complexes.

To investigate the impact of spatial positions of the transfer fork registration markers on the accuracy of generating a virtual dentofacial patient. An in vitro study was conducted using a mannequin head with a standard maxillary dentition model. Radiopaque gauge markers were fixed on the face and dentition of the mannequin head. CBCT was performed and the distance and angle between the dentition and facial markers were measured in the CBCT as reference values. Intraoral scanners were used to obtain 3D morphological data of the maxilla. Two types of transfer fork were designed and fabricated. The registration markers on transfer fork A were positioned in the midline area, while those on transfer fork B were located at the corners of the mouth on both sides. The transfer forks were digitised and connected to the maxillary dentition within the mannequin head, and facial scanning was performed using a facial scanner five times in each group. A virtual dentofacial patient was built through matching and integration of digital dentition, face and transfer fork data using 3D reverse engineering software (Geomagic Wrap 2021, 3D Systems, Rock Hill, SC, USA). Measurement values including feature lengths and feature angles between six facial gauge markers and three dentition gauge markers were obtained in the virtual patients. The mean trueness and precision of linear difference for virtual patients established using transfer fork A were -1.00 ± 0.11 mm and 0.27 ± 0.02 mm and the angle deviation was -1.88 ± 0.27 degrees, whereas for transfer fork B, the mean trueness and precision of linear difference were 2.66 ± 0.25 mm and 0.83 ± 0.06 mm, and the angle deviation was 3.74 ± 0.87 degrees. There is an overall significant difference in the trueness values of feature lengths (t = -13.963, P = 0.000) and angles (t = -5.985, P = 0.004) between transfer fork groups A and B, with group A showing better trueness and precision. Linear and angular errors will be introduced in the process of building up a virtual dentofacial patient using a transfer fork. The trueness and precision of the transfer fork with the matching markers at the centre of the lips are more precise than the transfer fork, with matching markers on both sides of the mouth.

Purpose Ensuring continued operation of industrial assets increasingly depends on replacing legacy components whose original tooling, suppliers and manufacturing routes are no longer available. Additive manufacturing (AM) offers a route to restore supply of such critical parts, but practical guidance on process selection and validation is limited. This study aims to establish and validate an end-to-end workflow for qualifying AM routes for legacy metal components in the energy sector, using an eccentric chemical pump gear as a representative case study. Design/methodology/approach A cast iron gear was reverse engineered using high-resolution 3D scanning and the geometry was refined through generative design to improve printability and reduce material usage. The part was reproduced using CoCr alloy by Laser Powder Bed Fusion (LPBF) and 4140 chromoly steel by sinter-based Bound Metal Deposition (BMD). Finite element analysis and LPBF process simulation were used to verify load-bearing performance and screen build orientations. Printed parts were postprocessed, inspected for dimensional accuracy, mechanically characterized and validated using API 677 gear contact testing and in-field operation. Findings Both AM routes achieved dimensional compliance and acceptable gear contact patterns, while exhibiting distinct trade-offs in mechanical response, accuracy, material consumption and workflow complexity. Successful in-field operation of the AM gear confirmed that the design-simulation-manufacture-test workflow can deliver functionally equivalent replacement parts and reduce dependence on obsolete casting routes. Originality/value A comparative, field-validated assessment is provided of LPBF and BMD for legacy component replacement in the energy sector. The study links reverse engineering, generative design, process simulation and experimental validation into an evidence-based process-selection template for qualifying AM routes for critical spare parts.

ABSTRACT The Kappel propeller has attracted considerable attention for its favorable hydrodynamic performance, yet further efficiency improvement under complex marine conditions remains challenging. This paper proposes a parameterized modeling and multi-objective optimization framework for Kap-type propeller. The F-spline technique is used to reconstruct and control the three-dimensional blade geometry, while point cloud-baded reverse engineering and experimental data are employed to validate the modeling and numercial simulation procedures. The results shows that the modified Kappel propeller improves propulsion efficiency by approximately 2.4%. Furthermore, an optimization framework coupling CFD with NSGA-II is established, with the thrust coefficient and open water efficiency as objective functions. Under the design condition, the optimized propeller increases open water efficiency by 3.62% and thrust coefficient by 6.51%. The proposed method effectively improves propulsive performance and provides a technicial reference for the parametric design and optimization of marine propellers.

Although abolishing the Crabtree effect in Saccharomyces cerevisiae through a pyruvate dehydrogenase bypass eliminates carbon loss through ethanol overflow metabolism, it compromises growth rates. While the Crabtree effect has been a valuable natural adaptation, it is energetically inferior to respiration and is generally undesirable in cell factories engineered to produce assimilatory compounds. Restoring growth efficiency in Crabtree-negative strains remains a central challenge. Through adaptive laboratory evolution of the engineered strain (sZJD23) and subsequent reverse engineering, a variant (sZJD28) with markedly improved growth was identified. This improvement is driven primarily by a mutation in MED2 (encoding a Mediator complex subunit) and, to a lesser extent, a mutation in GPD1 (encoding glycerol-3-phosphate dehydrogenase). By integrating quantitative proteomics with enzyme-constrained genome-scale modelling, we demonstrate that these mutations jointly enable a more efficient mode of oxidative stress adaptation and energy utilization. The GPD1 mutation suppresses a protein-costly, suboptimal NAD⁺-recycling strategy reliant on glycerol synthesis, while the MED2 mutation reshapes the oxidative stress response towards peroxisomal detoxification. Collectively, these adjustments optimize metabolic flux distribution and reduce protein costs in energy metabolism, thereby increasing ATP availability. Our findings reveal how coordinated mutations in regulatory and metabolic genes restore growth fitness in engineered Crabtree-negative yeast.

ABSTRACT Digital surveying has greatly improved the capture of complex historical structures, yet extracting meaningful information from 3D data remains time-consuming and inconsistent without automated methods, and existing workflows often rely on case-specific procedures. This contribution presents ReVault, an open-source Grasshopper script for Rhinoceros, offering a comprehensive pipeline to analyse 3D meshes of vaults, arches, domes, walls, pillars, buttresses and floors. The tool enables reverse engineering of original design principles, detection of construction traces and quantification of structural deformations. By jointly analysing vaults and their supports, deformation patterns can be identified that would otherwise remain hidden. The methods are demonstrated on four medieval buildings. Three comprehensive case studies combine vault and support analyses: at St John in Toulon-sur-Arroux (11th–12th century), nave widening created three hinge mechanisms in the dividing arches; at the Cathedral of St Peter in Lisieux (12th century), the southern façade leans outward by 69.3 cm with a previously unrecognized eastward component; and at the Basilica of St Anthony in Padua (13th century), arch opening is quantitatively linked to horizontal and vertical support deformation. The standardized, parametric workflow produces outputs suitable for long-term structural monitoring, HBIM modelling and structural simulation, advancing the documentation and preservation of heritage buildings.

Central nervous system injury causes motor deficits through derangement of excitatory, inhibitory, and/or neuromodulatory inputs to motoneurons, the three fundamental components of motor commands. Typically, study of pathologic neural control in humans is restricted to only one of the three. Chardon et al. (2024) presented a fundamentally new approach to comprehensively study all components by reverse engineering motor unit firing patterns. We apply their framework to motor unit firing patterns from 89 people with multiple sclerosis (MS) and 34 controls to study excitatory, inhibitory, and neuromodulatory contributions to pathologic motor output. Disruptions to all components are plausible in MS, a disease hallmarked by heterogeneity in nearly all aspects. Accordingly, we found abnormalities in MS for all three components. Notably, neuromodulation included both high and low extremes. Our results suggest that pathophysiology of motor commands in MS varies among patients, a finding fundamentally different from other studied populations showing relative consistency.

Recent advances in mitochondrial network dynamic and signalling highlight mitochondria as key therapeutic targets across diverse diseases. Yet, high drug development failure rates reflect an incomplete understanding of upstream molecular regulators of mitochondrial fate. Here, we address this gap by reverse engineering of the BH3-only protein BNIP3. Structural modelling and sequence–function analyses of its N-terminus identify a critical functional domain and amino acid hotspots that directly activate BCL-2 executioner proteins, triggering mitochondrial cell death. Leveraging these insights, we develop a BNIP3 antagonist peptide (B-017) that disrupts interactions between BNIP3 and BCL-2 executioner proteins, preserving mitochondrial integrity. B-017 demonstrates target specificity, a favourable safety profile, and robust suppression of cell death signalling in human cells. In clinically relevant animal models, it reduces tissue damage in the heart, brain, and liver. Together, these findings position B-017 as a promising therapeutic candidate targeting mitochondrial dysfunction.

Currently, there is an increasing demand for the treatment and reconstruction of interproximal papilla defects in the anterior aesthetic zone. However, the lack of standardized measurement methods and indicators for assessing anterior interproximal papilla defects undermines the scientific rigor and accuracy of related research, and hampers comparability across studies. Common measurement methods for interproximal papilla defects include direct periodontal probe measurement, intraoral photographic measurement, and intraoral scanning data measurement. The direct periodontal probe method is simple to perform with low accuracy yet. The intraoral photographic method, which involves taking intraoral images and analyzing with image analysis software, is susceptible to variations in shooting environment, parameter settings, and reference markers. The intraoral scanning data method, which combines intraoral scanning with reverse engineering software, offers advantages such as accuracy, reproducibility, and multi-dimensional assessment. However, it is influenced by scanning precision. This article reviews commonly used measurement methods for anterior interproximal papilla defects in recent years, aiming to provide research insights for relevant researchers and clinicians, promote the application of more accurate measurement methods in evaluating the severity of interproximal papilla defects, and ultimately improve the outcomes of papilla reconstruction procedures.

Quantitative and qualitative evaluation of precision in dental 3D facial scanning across handheld, fixed multicamera, and smartphone depth-sensing systems.

Multi-rotor plant protection drones are extensively utilized in rice field operations. However, the airflow generated by the rotors not only disturbs the rice canopy but also affects the deposition of pesticide droplets. The investigation of the impact of rotor wind fields on the protection of rice crops has the potential to enhance the effectiveness of operations of this nature. The present study firstly acquired aerial imagery of plant protection drone operations via the utilization of aerial photography, and subsequently employed machine vision technology to investigate the disturbance patterns of rotor wind fields on rice canopies. Notably, there is currently no effective experimental method to observe the disturbance of rice canopies caused by the wind field generated by plant protection drone rotors, and machine vision processing is proven to be an effective approach, which is a key innovation of this study. This study innovatively adopts an integrated experimental-numerical approach, a distinct advancement over existing Unmanned Aerial Vehicle (UAV) spraying and airflow research that typically focuses on single experimental or simulation methods. Aerial photography and machine vision were used to explore canopy disturbance patterns, while reverse engineering combined with thrust tests and numerical simulations verified rotor model precision and analyzed droplet deposition. Core findings show that flight speeds of 3-4m/s enable effective overlap between canopy disturbance and droplet deposition zones, achieving optimal plant protection effects; higher speeds cause droplet drift and zone misalignment, reducing efficacy. In addition, this study explores the spray deposition of plant protection drones based on numerical simulation methods, and assists in judging the effect of plant protection operations through the coincidence of droplet deposition and canopy disturbance zones. This integrated approach and related findings provide a reliable technical basis for optimizing UAV flight parameters, and have certain reference significance for the technical research and development of plant protection drones and the setting of operation parameters.

Ti-6Al-4 V dental implants are widely used because of their excellent biocompatibility and mechanical strength; however, their high elastic modulus relative to jawbone may cause stress shielding and peri-implant bone resorption. To improve biomechanical compatibility, this study introduced a gradient-regulated porous design into dental implants. A three-dimensional mandibular model was reconstructed from CT image data using reverse engineering techniques. Based on Gibson-Ashby theory, several porous architectures with a unified external octet-truss framework but different internal topologies were established, including orthogonally intersecting plates (structure 1), face-diagonal oblique plates (structure 2), an inner octahedral topology (structure 3), and the original octet-truss structure as the reference configuration. At a fixed porosity of 65%, the effects of key geometric parameters, including strut diameter and hole diameter, on the equivalent elastic modulus and bending response of these porous structures were systematically investigated by finite element analysis, and axial compression and bending simulations were combined to screen candidate structures. The results showed that structure 3 exhibited the lowest equivalent elastic modulus (16.92-19.42 GPa). In particular, structure 3(c), with a strut diameter of 0.12 mm and a hole diameter of 0.2 mm, achieved an equivalent elastic modulus of 17.00 GPa, which was 21.73% lower than that of the original octet-truss structure (21.72 GPa). Under bending loading, its magnitude of displacement was 0.014 mm, close to that of the original octet-truss structure (0.0138 mm), indicating that stiffness reduction was achieved without an obvious loss of bending stability. To further address the limited geometric adaptability of conventional regular lattices, a unit-cell-based conformal filling strategy was proposed to generate a plate-lattice porous implant matching the contour of a 5 mm diameter implant. Finally, analysis under clinically relevant occlusal loading conditions showed that the porous implant model exhibited a favorable stress-transfer pattern, promoted a relatively uniform stress distribution in the surrounding bone, and improved displacement compatibility at the bone-implant interface. These findings provide theoretical support for the structural design of porous dental implants and offer a useful basis for further structural optimization and biomechanical adaptation.

Towards the automated extraction and refactoring of NoSQL schemas from application code

3D-printed skeletal tissue analogues using bone decellularized extracellular Matrix with PCL-bioglass composite: A biomimetic approach.

L-valine is an essential amino acid for animal nutrition. Ideally, it can be produced from D-glucose through homotypic L-valine fermentation in a growth-coupled manner. To date, no known microorganism, native or engineered, can grow on D-glucose and ammonia anaerobically with L-valine as the sole product. Here, we direct the metabolic flux through a reinforced L-valine synthetic pathway by blocking mixed-acid fermentation and L-alanine synthesis reactions to create an NADH driving force in Escherichia coli. We further evolve the engineered strain to debottleneck growth constraints by anaerobic growth rescue. The resulting evolved hyper-valine producer converts D-glucose in a 320 m3 reactor to 83.6 g/L L-valine within 60 h, reaching a yield of 0.55 g/g glucose (85% of the theoretical maximum). Through reverse engineering, we identify that more than a 10-fold improvement in anaerobic growth and L-valine production rate arises from the amplified L-valine synthetic pathway, the additional electron sinks and reprogramming of global regulation. Together, we changed the way of L-valine production into homotypic L-valine fermentation and demonstrate how E. coli variants adapted their metabolic activities and transcriptional regulation to boost fitness in an anoxic condition, with L-valine synthesis serving as the primary NADH-consuming pathway.

ABSTRACT In recent years, Terrestrial Laser Scanner (TLS) is widely used in urban planning, cultural heritage protection, and many other aspects in the field of architecture. However, the point cloud model obtained by TLS has some problems, such as unstable accuracy, missing key points, which seriously affect the practical application and popularization of TLS technology. According to ASTM E3125 issued by American Society for Testing and Materials, TLS is affected by several field variables such as scanning distance and scanning angle. At the same time, we also know that the relative position between scanning points will also affect the accuracy of point cloud model by the actual measurement experience. Therefore, this paper selects scanning distance, scanning angle, and relative position between scanning points as the three field controllable variables, design univariate and multivariate compound experiments to discuss the influence on the accuracy in detail. A precise control range of 4 m scanning spacing, 4 m scanning point distance, and 71° scanning angle was obtained. The rationality of the conclusion proposed in this paper is verified in the actual measurement project, which contributes to improving the accuracy of TLS operations.

Aromatics are important building blocks for polymers, pharmaceuticals, and advanced materials, but their current production relies on petrochemical processes. Biotechnological de novo production from renewable bio-based feedstocks with microbial cell factories provides a sustainable alternative. In this study, we enhanced 4-coumarate production in Pseudomonas taiwanensis from glucose and glycerol compared to previously published producers. This was achieved through heterologous expression of tyrosine ammonia-lyase (TAL) from Rivularia sp. PCC7116, which debottlenecked the specific deamination of tyrosine. Moreover, deletion of the phosphoenolpyruvate carboxylase-encoding gene ppc further increased the production. Subsequently, the substrate spectrum for efficient aromatics production was expanded to include the abundant pentoses, xylose and arabinose. Heterologous non-oxidative assimilation pathways were integrated into P. taiwanensis GRC3 chassis strains and growth on xylose and arabinose was improved through adaptive laboratory evolution, whole-genome sequencing, and reverse engineering. Optimized catabolic modules were then transferred to producer strains to enhance or enable 4-coumarate production from xylose and arabinose. Notably, the product yield on xylose increased approximately 3.5-fold with the non-oxidative xylose isomerase pathway compared to the oxidative native Weimberg pathway, without compromising yields on glucose. For the final strain, P. taiwanensis GRC3Δ6-TYR2Δppc-REXA-attTn7::P14f-RpcTAL, product yields were significantly higher on xylose (38.2% (Cmol/Cmol)) and arabinose (39.7% (Cmol/Cmol)) than on glucose (26.0% (Cmol/Cmol)). 4-Coumarate production was characterized on mixtures of glucose, xylose, and arabinose to mimic lignocellulosic hydrolysate feedstocks, with the best reverse-engineered xylose- and arabinose-metabolizing 4-coumarate producer significantly outperforming the reference strain.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12934-026-03014-w.

This in vitro study aimed to evaluate the regional trueness and precision of three scanning protocols for digitizing complete denture prototypes that have undergone border molding and wash impression procedures. Maxillary and mandibular edentulous typodonts were scanned and used to fabricate complete denture prototypes. The complete denture prototypes were additively manufactured using try-in resin. Both prototypes were utilized to obtain final impressions using polyvinylsiloxane material. Scans of the washed prototypes were made using a desktop E4 scanner (3Shape A/S) to serve as the reference control. Test scans were acquired using an intraoral scanner TRIOS4 (3Shape A/S). Three scanning protocols were employed to digitize the washed prototypes: Intaglio-First (IF), Occlusal-First (OF), and Peripheral Frame (PF), with 10 scans per group (n = 10). Reference meshes were segmented into three regions of clinical interest: overall prosthesis, denture-bearing area (DBA), and dentition. Test scans were superimposed onto reference scans using best-fit alignment in reverse engineering software. Trueness was quantified as the root mean square (RMS) deviation, and precision was assessed as the standard deviation of RMS values among repeated scans. A one-way analysis of variance (ANOVA) with post hoc comparison was used to analyze the data (α = 0.05). For the maxillary arch, the IF protocol demonstrated significantly superior trueness compared to the PF protocol for the dentition (p = 0.0147) and overall prosthesis (p = 0.0053) regions. The IF and OF protocols also showed significantly higher precision than the PF protocol for maxillary dentition and overall prosthesis regions. No significant differences among protocols were found for the maxillary DBA. For the mandibular arch, no significant differences in trueness or precision were observed among protocols for any region. Trueness values across all protocols and regions ranged from 0.048 to 0.088mm. Scanning protocol significantly influenced the trueness and precision of maxillary digitized prototypes, with the IF protocol demonstrating superior accuracy. However, this did not hold true for mandibular prototypes. All scanning protocols produced accuracy values that were clinically acceptable for complete denture fabrication, supporting the clinical viability of intraoral scanning for digitizing border-molded denture prototypes regardless of scanning protocol selection.

The 2D dendritic structure of stream networks within fluvial catchments has not previously been linked to reach-scale channel processes. Existing extensions of 2D network models to 3D landscapes yield only pixelized landscapes, without resolving channels or their properties. At-a-point relations for hydraulic geometry alone provide no insight into how fluvial processes sculpt 3D landscapes in which channels are embedded. We provide this insight by reverse-engineering generic fluvial landscapes with scalable dimensions for all attributes. We do this by coupling a) a 2D network generator, b) dimensionless, physically grounded relations for gravel-bed river hydraulic geometry, and c) a simplified hillslope model. Catchment hypsometric curve and relief, as well as relations between local channel properties and basin structure, cannot be predicted with just one of these components. Our model predicts in specific, dimensioned terms not only how attributes such as hypsometric curve change with changing bed grain size, precipitation and catchment area, but also why, because each model element can be interrogated individually or jointly. Our method offers a tool for studying the effect of varied mean annual precipitation and intensity on catchment structure. Our central application is to tectonically inactive, quasi-equilibrium, low-relief montane landscapes on Earth. We underline the insight provided by our formulation through application to analogous planetary fluvial landscapes. We implement this for Mars and Titan using appropriate values for gravitational acceleration and sediment submerged specific gravity, both freely variable in our model. Our reverse-engineering methodology indicates why and how an analogous class of landscapes should differ in different planetary settings.