Extensive Steps Research Articles

The β-Ketoacyl-ACP Synthase FabF Catalyzes Carbon-Carbon Bond Formation in a Bimodal Pattern for Fatty Acid Biosynthesis.

Fatty acids produced by the type-II fatty acid biosynthetic pathway (FAS-II) are essential biomaterials for bacterial membrane construction and numerous metabolic routes. The β-ketoacyl-ACP synthase FabF catalyzes the key C-C bond formation step for fatty acid elongation in FAS-II. Here, we revealed the substrate recognition and catalytic mechanisms of FabF by determining FabF-ACP complexes. FabF displays a distinctive bimodal catalytic pattern specifically on C6 and C10 acyl-ACP substrates. It utilizes positively charged residues located on the η3-helix and loop1 regions near the catalytic tunnel entrance to bind ACP, and two hydrophobic cavities as well as "front", "middle", and "back" door residues to specifically stabilize C6 and C10 acyl substrates for preferential catalysis. Further quantum chemistry calculations suggest that the FabF catalytic residues Lys336 and His304 facilitate proton transfer during condensation catalysis and C-C bond formation. Our results provide key mechanistic insights into the biosynthesis of molecular carbon skeletons based on ketosynthases that are highly conserved through the FAS and polyketide synthase (PKS) analogous biosynthetic routes, broaden the understanding of the tricarboxylic acid cycle that utilizes lipoic acid derived from C8-ACP accumulated due to the FabF distinctive catalytic pattern for oxidative decarboxylations, and may facilitate the development of narrow-spectrum antibacterial drugs.

Divergent Tandem Acyl Carrier Proteins Necessitate In-Series Polyketide Processing in the Leinamycin Family.

The leinamycin family of polyketides are promising antitumour antibiotics, yet several aspects of their biosynthesis remain elusive. All leinamycin family members bear a sulfur-containing moiety which is essential for the anticancer activity exhibited by leinamycin. The key building blocks required for the incorporation of these functionalities are introduced in the final module of the polyketide synthase (PKS), which elegantly combines β-branching and thiocysteine incorporation to generate a diverse library of sulfur-based molecular scaffolds. Two acyl carrier proteins (ACPs) form a key didomain component of this module, but their amino acid sequence divergence has brought into question the common notion of functional equivalence. Here, we provide unprecedented functional evidence that these tandem ACPs play distinct roles in the final module of polyketide assembly. Using the weishanmycin biosynthetic pathway as a template, the in vitro reconstitution of key polyketide chain extension and β-branching steps in this module has revealed strict functional selectivity for a single ACP. Furthermore, we propose a cryptic transacylation step must occur prior to polyketide off-loading and cyclization. Altogether, these mechanistic investigations suggest that an atypical in-series mechanism underpins sulfur incorporation in the leinamycin family, and provides significant progress towards delineating their late-stage assembly.

Biochemically Programmable Isothermal PCR.

Isothermal PCR can be performed by imposing a static temperature gradient that continuously circulates reagents through denaturing, annealing, and extension conditions inside a PCR tube. But despite early promise, these systems have yet to demonstrate performance and repeatability sufficient for adoption in validated laboratory tests because the rate-limiting extension step is inherently short and cannot be increased independently of the other stages in a temperature cycle. Here, a discovery that enables isothermal PCR to be achieved with statistically robust repeatability that meets or exceeds diagnostic assay requirements (false positive/negative rate <8% at 95% confidence) by manipulating the interplay between the DNA replication biochemistry (via the amplicon GC content) and the microscale circulatory flow inside a PCR tube is reported. Surprisingly, optimal performance depends on selecting primer sequences that replicate high GC content amplicons, contradicting established PCR primer design rules. This innovative thermocycling approach accelerates PCR to speeds rivaling ultra-fast instruments, enabling rapid, repeatable isothermal DNA analysis across a range of targets relevant to diagnostics and pathogen detection.

Formaldehyde quantification using gas chromatography–mass spectrometry reveals high background environmental formaldehyde levels

Formaldehyde (HCHO) is a human toxin that is both a pollutant and endogenous metabolite. HCHO concentrations in human biological samples are reported in the micromolar range; however, accurate quantification is compromised by a paucity of sensitive analysis methods. To address this issue, we previously reported a novel SPME–GC–MS-based HCHO detection method using cysteamine as an HCHO scavenger. This method showed cysteamine to be a more efficient scavenger than the widely used O-(2,3,4,5,6-pentafluorobenzyl)hydroxylamine, and enabled detection of aqueous HCHO in the nanomolar range and quantification in the micromolar range. However, quantification in this range required immersive extraction of the HCHO-derived thiazolidine, while a high background signal was also observed. Following on from these studies, we now report an optimised head-space extraction SPME–GC–MS method using cysteamine, which provides similarly sensitive HCHO quantification to the immersive method but avoids extensive wash steps and is therefore more amenable to screening applications. However, high background HCHO levels were still observed A Complementary GC–MS analyses using a 2-aza-Cope-based HCHO scavenger also revealed high background HCHO levels; therefore, the combined results suggest that HCHO exists in high (i.e. micromolar) concentration in aqueous samples that precludes accurate quantification below the micromolar range. This observation has important implications for ongoing HCHO quantification studies in water, including in biological samples.

Efficient electrochemical nitrate reduction by iron phthalocyanine nanorods decorated on multi-walled carbon nanotubes

Electrochemical nitrate reduction reaction (NO3RR) has the dual effect of relieving nitrate contamination and producing the high value-added NH3, which calls for the creation of sophisticated electrodes in order to attain high yield rate, stability, and Faradaic efficiency (FENH3). In this study, iron-based conjugated molecule electrocatalyst, i.e. iron phthalocyanine (FePc), was fabricated into nanorods with excellent crystallinity by a facile precipitation method and applied in NO3RR electrode. The NO3RR performance is enhanced by decorating FePc nanorods on multi-walled carbon nanotubes (CNT) (denoted as FePc@CNT X:10) due to the improved dispersibility and electrical conductivity. The results disclose that FePc@CNT 6:10 exhibits the highest performance with a FENH3 of 94% at −0.45 V (vs. Reversible Hydrogen Electrode) and an ammonia yield rate of 0.35 mmol h−1 mgcat.−1. Notably, a significant NH3 partial current density of approximately 75 mA cm−2 is attained at −0.50 V, correlating with a marginally deteriorated FENH3 of 92.9% and an ammonia yield rate of 0.46 mmol h−1 mgcat.−1. Besides, this catalyst exhibits a long-term NO3RR stability with the ten consecutive cycles. Operando attenuated total reflection surface-enhanced infrared absorptive spectroscopy (ATR-SEIRAS) and density functional theory (DFT) calculations reveal that NO3− undergoes deoxygenation and extensive hydrogenation steps and FePc is more likely to convert NO3− to NH3 under the ONH route. This work provides advanced insights into the rational design of metal phthalocyanine catalysts for electrocatalytic reactions.

Passive Wireless Porous Biopolymer Sensors for At-Home Monitoring of Oil and Fatty Acid Nutrition.

Dietary oils─rich in omega-3, -6, and -9 fatty acids─exhibit critical impacts on health parameters such as cardiovascular function, bodily inflammation, and neurological development. There has emerged a need for low-cost, accessible method to assess dietary oil consumption and its health implications. Existing methods typically require specialized, complex equipment and extensive sample preparation steps, rendering them unsuitable for home use. Addressing this gap, herein, we study passive wireless, biocompatible biosensors that can be used to monitor dietary oils directly from foods either prepared or cooked in oil. This design uses broad-coupled split ring resonators interceded with porous silk fibroin biopolymer (requiring only food-safe materials, such as aluminum foil and biopolymer). These porous biopolymer films absorb oils at rates proportional to their viscosity/fatty acid composition and whose response can be measured wirelessly without any microelectronic components touching food. The engineering and mechanism of such sensors are explored, alongside their ability to measure the oil presence and fatty acid content directly from foods. Its simplicity, portability, and inexpensiveness are ideal for emerging needs in precision nutrition─such sensors may empower individuals to make informed dietary decisions based on direct-from-food measurements.

Landfill reclamation using the example of the municipal waste disposal plant in Puławy – a case study

Landfilling remains a dominant method of waste disposal worldwide, largely due to inadequate waste management frameworks, despite efforts to promote recycling and reuse. As populations and consumption levels grow, waste generation continues to increase, highlighting the need for effective waste management solutions. Landfills are sites where waste undergoes various physical, chemical and biological transformations, making the implementation of safety measures, leachate collection and gas recovery systems critical. This article presents a case study of the reclamation of the Puławy municipal landfill, which has been in operation since 1998. The reclamation process involved extensive steps such as forming the landfill body, securing the slopes, constructing a degassing layer, implementing a sealing layer and applying a cover layer. These measures were crucial in mitigating the negative environmental impact of the landfill, reducing harmful emissions and improving local groundwater quality. The results of the Puławy municipal landfill reclamation include reductions in leachate and gas emissions, improvements in air/groundwater quality and the transformation of the site into a green space that provides recreational opportunities for the community. This reclamation project serves as a model of responsible waste management and community involvement, demonstrating how degraded landfills can be transformed into valuable public resources. The article also discusses the broader implications of landfill reclamation, noting both the positive and negative aspects. Benefits include environmental protection, resource recovery, enhanced biodiversity and improved quality of life for local communities. However, challenges such as the long-term monitoring and maintenance of reclaimed sites and the financial costs associated with reclamation efforts are also considered. This study also underscores the importance of landfill reclamation in the process of shaping and promoting environmental sustainability, as well as addressing modern waste management challenges.

Discrete Cosine Transform Algorithm Analysis of The Application of QR Codes on Birth Certificate Image Objects

Security and authentication of important documents such as birth certificates have become a major concern today and in the digital age. In this context, the use of watermarking technology is an effective solution to add additional information to digital documents without destroying their authenticity. The DCT method is used to convert images into the frequency domain, allowing the watermark to be added without disturbing the appearance of the document. Testing ensures that the watermarking process does not damage the appearance of the original document and that the watermark information can be accurately re-extracted. The research results show that the QR code watermark using the DCT method successfully provides secure and authentic birth certificates. The application of the Discrete Cosine Transform (DCT) method to the birth certificate QR code watermarking process was able to significantly improve document security and authentication without compromising document quality or important information. This thesis outlines the extensive steps in the practical application of the QR code watermark on birth certificates. From data preparation including birth certificate images and QR codes to the watermark conversion and extraction process, all steps are explained in detail. Thus, applying this watermarking technique can be an effective solution to protect the integrity of birth certificates and similar documents in the digital environment.

Pore structure and N/F bifunctional groups in hierarchical porous carbons via spontaneous silica etching for enhanced capacitive deionization performance

This study presents a novel, one-step synthesis of nitrogen and fluorine bifunctional hierarchical porous carbons (HPCs) for capacitive deionization (CDI). The method involves co-gelation of silica and polyvinylidene fluoride, followed by pyrolysis for efficient silica removal. This method eliminates the need for toxic solvents and extensive cleaning steps. Optimized HPCs exhibit a high surface area (1464.71 m2/g) with abundant mesopores. Electrochemical analysis reveals a specific capacitance of 179.5F/g. In the CDI process using 500 mg/L NaCl solution, the HPCs demonstrate exceptional performance with a high salt adsorption capacity (24.61 mg/g) and a fast average adsorption rate (3.69 mg/g/min). This performance is attributed to the combined effects of the hierarchical pore structure facilitating rapid ion adsorption and the N/F bifunctional groups enhancing electrosorption ability. This work offers a one-step, solvent-free method for synthesizing HPCs with superior desalination performance and demonstrates the synergistic effect of pore structure and N/F bifunctional groups.

Mitochondrial RNA Helicases: Key Players in the Regulation of Plant Organellar RNA Splicing and Gene Expression.

Mitochondrial genomes of land plants are large and exhibit a complex mode of gene organization and expression, particularly at the post-transcriptional level. The primary organellar transcripts in plants undergo extensive maturation steps, including endo- and/or exo-nucleolytic cleavage, RNA-base modifications (mostly C-to-U deaminations) and both 'cis'- and 'trans'-splicing events. These essential processing steps rely on the activities of a large set of nuclear-encoded factors. RNA helicases serve as key players in RNA metabolism, participating in the regulation of transcription, mRNA processing and translation. They unwind RNA secondary structures and facilitate the formation of ribonucleoprotein complexes crucial for various stages of gene expression. Furthermore, RNA helicases are involved in RNA metabolism by modulating pre-mRNA maturation, transport and degradation processes. These enzymes are, therefore, pivotal in RNA quality-control mechanisms, ensuring the fidelity and efficiency of RNA processing and turnover in plant mitochondria. This review summarizes the significant roles played by helicases in regulating the highly dynamic processes of mitochondrial transcription, RNA processing and translation in plants. We further discuss recent advancements in understanding how dysregulation of mitochondrial RNA helicases affects the splicing of organellar genes, leading to respiratory dysfunctions, and consequently, altered growth, development and physiology of land plants.

NKX2-2 based nuclei sorting on frozen human archival pancreas enables the enrichment of islet endocrine populations for single-nucleus RNA sequencing

BackgroundCurrent approaches to profile the single-cell transcriptomics of human pancreatic endocrine cells almost exclusively rely on freshly isolated islets. However, human islets are limited in availability. Furthermore, the extensive processing steps during islet isolation and subsequent single cell dissolution might alter gene expressions. In this work, we report the development of a single-nucleus RNA sequencing (snRNA-seq) approach with targeted islet cell enrichment for endocrine-population focused transcriptomic profiling using frozen archival pancreatic tissues without islet isolation.ResultsWe cross-compared five nuclei isolation protocols and selected the citric acid method as the best strategy to isolate nuclei with high RNA integrity and low cytoplasmic contamination from frozen archival human pancreata. We innovated fluorescence-activated nuclei sorting based on the positive signal of NKX2-2 antibody to enrich nuclei of the endocrine population from the entire nuclei pool of the pancreas. Our sample preparation procedure generated high-quality single-nucleus gene-expression libraries while preserving the endocrine population diversity. In comparison with single-cell RNA sequencing (scRNA-seq) library generated with live cells from freshly isolated human islets, the snRNA-seq library displayed comparable endocrine cellular composition and cell type signature gene expression. However, between these two types of libraries, differential enrichments of transcripts belonging to different functional classes could be observed.ConclusionsOur work fills a technological gap and helps to unleash frozen archival pancreatic tissues for molecular profiling targeting the endocrine population. This study opens doors to retrospective mappings of endocrine cell dynamics in pancreatic tissues of complex histopathology. We expect that our protocol is applicable to enrich nuclei for transcriptomics studies from various populations in different types of frozen archival tissues.

Achieving High-Quality Text and Audio-to-Image Generation in a Single Step

Diffusion models have significantly advanced text-to-image generation by producing high-quality and imaginative results. However, their multi-step sampling process often proves slow, requiring extensive inference steps to achieve satisfactory outcomes. Despite attempts to improve sampling speed and computational efficiency through distillation, creating a functional one-step model has remained elusive. In this study, we investigate Rectified Flow, a recent method primarily applied to small datasets, as a potential solution. Central to Rectified Flow is its reflow procedure, which optimizes probability flow trajectories, refines noise-to- image mapping, and enables effective distillation with student models. We introduce a novel text-conditioned pipeline to convert Stable Diffusion (SD) into an ultra-fast one-step model. Our approach underscores the crucial role of reflow in enhancing noise-to-image assignments. Leveraging this pipeline, we develop the first one-step diffusion-based text-to- image generator capable of producing high-quality images comparable to those generated by SD. Additionally, we extend our methodology to include audio inputs, demonstrating its efficacy in generating images from audio cues with remarkable fidelity and speed. Key Words: Stable Diffusion, Text-to-Image Creation, Image Processing

Towards a smart PCR process

ABSTRACT PCR has transformed DNA profiling, providing the opportunity to analyse trace DNA. Trace DNA often contains degraded DNA, such that many samples generate no meaningful STR data. The PCR process has changed little since first use despite the drive for greater sensitivity: the same steps of denaturation, annealing and extension used in the first STR amplifications are recognizable in current processes. Our aim is to move towards a smart PCR that can monitor amplification and amend cycling conditions. To achieve this, DNA amplification needs to be performed on a real-time PCR machine. Initially, this was the adjustment of the time at the denaturation and annealing steps. Standard STR profiling was performed using the GlobalFilerTM kit on an open-source qPCR machine and reduced in step changes the denaturation time to 2 s from 10 and time of annealing from 90 s to 30. The comparison of the STR profiles resulting from the modified amplification parameter to those of the control set indicated a loss of allelic amplification although the resulting likelihood ratios were extremely high. We identify the groundwork required to attain the ambitious goal of creating a smart PCR system that can respond in real time.

Quad-Rotor Unmanned Aerial Vehicle Path Planning Based on the Target Bias Extension and Dynamic Step Size RRT* Algorithm

For the path planning of quad-rotor UAVs, the traditional RRT* algorithm has weak exploration ability, low planning efficiency, and a poor planning effect. A TD-RRT* algorithm based on target bias expansion and dynamic step size is proposed herein. First, random-tree expansion is combined with the target bias strategy to remove the blindness of the random tree, and we assign different weights to the sampling point and the target point so that the target point can be quickly approached and the search speed can be improved. Then, the dynamic step size is introduced to speed up the search speed, effectively solving the problem of invalid expansion in the process of trajectory generation. We then adjust the step length required for the expansion tree and obstacles in real time, solve the opposition between smoothness and real time in path planning, and improve the algorithm’s search efficiency. Finally, the cubic B-spline interpolation method is used to modify the local inflection point of the path of the improved RRT* algorithm to smooth the path. The simulation results show that compared with the traditional RRT* algorithm, the number of iterations of path planning of the TD-RRT* algorithm is reduced, the travel distance from the starting position to the end position is shortened, the time consumption is reduced, the path route is smoother, and the path optimization effect is better. The TD-RRT* algorithm based on target bias expansion and dynamic step size significantly improves the planning efficiency and planning effect of quad-rotor UAVs in a three-dimensional-space environment.

Determination of Anchor Drop Sequence during Vessel Anchoring Operations Based on Expert Knowledge Base and Hydrometeorological Conditions

Presently, the most common technique for maintaining a ship’s location is dynamic positioning, which uses a series of thrusters to hold its position. This method is resilient to moderate hydro-meteorological conditions, eliminating the need for extensive preliminary steps before initiating positioning operations. An alternative approach involves station keeping using a set of anchors, where thrusters are not employed, necessitating careful planning of the anchorage in light of hydro-meteorological conditions. Presently, in vessels using this anchoring method, the captain determines the order of anchor drops, taking into account the prevailing weather conditions, the ship’s maneuvering abilities, and vessel capability plots. This article introduces a novel algorithm that uses sensor-acquired weather data and a cognitive knowledge base to establish the best sequence for anchor drops. This innovation represents a significant stride towards the automation of the anchoring process. By using the anchorage planning algorithm presented in this publication, it has been possible to reduce the time required for anchor deployment by about 52%, due to the preparation of the anchor deployment strategy in port. A reduction in energy consumption of about 8% was also achieved.

Transformative Approach for Heart Rate Prediction from Face Videos Using Local and Global Multi-Head Self-Attention

Heart rate estimation from face videos is an emerging technology that offers numerous potential applications in healthcare and human–computer interaction. However, most of the existing approaches often overlook the importance of long-range spatiotemporal dependencies, which is essential for robust measurement of heart rate prediction. Additionally, they involve extensive pre-processing steps to enhance the prediction accuracy, resulting in high computational complexity. In this paper, we propose an innovative solution called LGTransPPG. This end-to-end transformer-based framework eliminates the need for pre-processing steps while achieving improved efficiency and accuracy. LGTransPPG incorporates local and global aggregation techniques to capture fine-grained facial features and contextual information. By leveraging the power of transformers, our framework can effectively model long-range dependencies and temporal dynamics, enhancing the heart rate prediction process. The proposed approach is evaluated on three publicly available datasets, demonstrating its robustness and generalizability. Furthermore, we achieved a high Pearson correlation coefficient (PCC) value of 0.88, indicating its superior efficiency and accuracy between the predicted and actual heart rate values.

From slides to insights: Harnessing deep learning for prognostic survival prediction in human colorectal cancer histology.

Prognostic survival prediction in colorectal cancer (CRC) plays a crucial role in guiding treatment decisions and improving patient outcomes. In this research, we explore the application of deep learning techniques to predict survival outcomes based on histopathological images of human colorectal cancer. We present a retrospective multicenter study utilizing a dataset of 100,000 nonoverlapping image patches from hematoxylin & eosin-stained histological images of CRC and normal tissue. The dataset includes diverse tissue classes such as adipose, background, debris, lymphocytes, mucus, smooth muscle, normal colon mucosa, cancer-associated stroma, and colorectal adenocarcinoma epithelium. To perform survival prediction, we employ various deep learning architectures, including convolutional neural network, DenseNet201, InceptionResNetV2, VGG16, VGG19, and Xception. These architectures are trained on the dataset using a multicenter retrospective analysis approach. Extensive preprocessing steps are undertaken, including image normalization using Macenko's method and data augmentation techniques, to optimize model performance. The experimental findings reveal promising results, demonstrating the effectiveness of deep learning models in prognostic survival prediction. Our models achieve high accuracy, precision, recall, and validation metrics, showcasing their ability to capture relevant histological patterns associated with prognosis. Visualization techniques are employed to interpret the models' decision-making process, highlighting important features and regions contributing to survival predictions. The implications of this research are manifold. The accurate prediction of survival outcomes in CRC can aid in personalized medicine and clinical decision-making, facilitating tailored treatment plans for individual patients. The identification of important histological features and biomarkers provides valuable insights into disease mechanisms and may lead to the discovery of novel prognostic indicators. The transparency and explainability of the models enhance trust and acceptance, fostering their integration into clinical practice. Research demonstrates the potential of deep learning models for prognostic survival prediction in human colorectal cancer histology. The findings contribute to the understanding of disease progression and offer practical applications in personalized medicine. By harnessing the power of deep learning and histopathological analysis, we pave the way for improved patient care, clinical decision support, and advancements in prognostic prediction in CRC.

Regression Algorithms in Predicting the SARS-CoV-2 Replicase Polyprotein 1ab Inhibitor: A Comparative Study

Due to its extensive steps and trials, drug discovery is a long and expensive process. In the last decade, as also hard pressed by the COVID-19 pandemic, the screening process could be assisted with the advancement in computational technology including the application of Machine Learning. The classification task in Machine Learning has become one of the major approaches for drug discovery. Unfortunately, this practice uses discretized labels that might lead to the loss of quantitative properties that could be meaningful. Therefore, in this paper, we aim to compare various Machine Learning regression algorithms in predicting inhibitory bioactivity, specifically the IC50 value, with the SARS-CoV-2 Replicase Polyprotein 1ab as the target. With 1,138 non-duplicated data downloaded from the ChEMBL database that was engineered into four dataset variances, 42 regression algorithms were utilized for the prediction. We found that there are computational challenges to the use of regression algorithms in predicting bioactivity, for only a handful and a specific dataset variance that returned valid performance parameters upon testing. The three that yielded the highest counts of valid performance parameters are the Histogram Gradient Boosting Regressor (HGBR), Light Gradient Boosting Machine Regressor (LGBR), and Random Forest Regression (RFR). Further statistical analyses show that there is no significant difference between these three algorithms, except for the time taken for training and testing the model, where the LGBR excels. Therefore, these three algorithms should be primarily considered for the study with the same nature.

Experimental procedures to investigate fibrillation of proteins

The unwanted phenomenon of protein fibrillation is observed in vivo and during therapeutic protein development in the industry. Protein aggregation is associated with various degenerative disorders and might induce immune-related challenges post-administration of biopharmaceutics. A pipeline for early detection, identification, and removal of pre-formed fibrils is needed to improve the quality, efficacy, and effectiveness of the formulation. Protein fibril formation is accompanied by unfolding, secondary structural changes and the formation of larger aggregates. However, most detection processes come with extensive sample preparation steps and inefficient repeatability, incurring a financial burden on research. The current article summarizes and critically discusses six simple yet powerful methods to detect aggregation phenomena in the line of detecting fibrillar aggregates in heat-induced bovine serum albumin protein. Comparing the native and heat-induced protein samples would provide insights about aggregates. Easy, inexpensive and optimized protocols for detecting the fibrillation of proteins are explained. The procedures mentioned here detected the appearance of β-sheet-rich fibrils in the heat-induced protein sample. The aggregation is characterized by enhanced thioflavin-T fluorescence, alteration in the intrinsic fluorescence, decrease in helicity and subsequent increase in β-sheet and appearance of particles with larger hydrodynamic diameters.•This article summarizes various analytical techniques to easily characterize the fibrillation of proteins.•Various techniques to detect the formation of β-sheet rich structures, changes in the secondary structures and size of aggregates have been discussed.•The stated methodologies are validated on a model protein, albumin.

Measuring vertical track irregularities from instrumented heavy haul railway vehicle data using machine learning

This paper proposes a data-driven approach to estimating geometric track irregularities from instrumented railway vehicle (IRV) data. Machine learning is used to find the nonlinear mapping between IRV data and track irregularities. A dynamic model of the BRA1 railway vehicle was used to generate an artificial dataset that contains variables that are measured by the real BRA1 IRV and other variables measured by IRVs found in the literature. An extensive data analysis step was done to verify if the current instrumentation of the BRA1 IRV is sufficient for obtaining both lateral and vertical track irregularities. Feature engineering based on wagon movements, signal integration and time domain statistical metrics were applied to extract features and then the best features were selected using a wrapper method. Eight different regression ML models were trained and optimized after the feature selection using Optuna. The results show that, with the current instrumentation of the BRA1 IRV, obtaining lateral track irregularities is unlikely due to low correlation, however, vertical irregularities can be obtained with a root mean squared error (RMSE) of 0.556 mm. With postprocessing, the RMSE was further reduced to 0.410 mm.