Large Datasets Research Articles

Advances in rock physics for pore pressure prediction: A comprehensive review and future directions

Advances in rock physics have significantly enhanced pore pressure prediction, a critical aspect of subsurface exploration and drilling operations. This comprehensive review delves into the latest developments in rock physics methodologies, integrating empirical, theoretical, and computational approaches to predict pore pressure more accurately. Traditional pore pressure prediction methods often rely on well log data and seismic attributes, but recent advancements have introduced innovative techniques that leverage the physical properties of rocks to provide more reliable predictions. Key advances include the development of improved rock physics models that better account for the complexities of subsurface environments, such as heterogeneity and anisotropy. These models integrate data from various sources, including well logs, core samples, and seismic surveys, to create a more comprehensive understanding of the subsurface. Additionally, the application of machine learning and artificial intelligence to rock physics has opened new avenues for analyzing large datasets, identifying patterns, and refining predictive models. This review also examines the role of laboratory experiments and field studies in validating and calibrating rock physics models. High-pressure and high-temperature experiments have provided valuable insights into the behavior of rocks under different conditions, which are essential for accurate pore pressure prediction. Field studies, on the other hand, offer real-world data that help in fine-tuning models and methodologies. Future directions in rock physics for pore pressure prediction include the integration of advanced geophysical techniques, such as full-waveform inversion and distributed acoustic sensing, which offer higher resolution data and more detailed subsurface imaging. The use of cloud computing and high-performance computing platforms is also expected to enhance the processing and analysis of large datasets, making predictive models more efficient and scalable. The comprehensive review concludes by highlighting the importance of interdisciplinary collaboration in advancing rock physics methodologies. By combining expertise from geophysics, petrophysics, geomechanics, and data science, the field can continue to innovate and improve the accuracy and reliability of pore pressure predictions, ultimately enhancing exploration and production efficiency in the oil and gas industry. Keywords: Advances, Rock Physics, Pore Pressure, Prediction, Future Directions.

Multimodal soft valve enables physical responsiveness for preemptive resilience of soft robots.

Resilience is crucial for the self-preservation of biological systems: Humans recover from wounds thanks to an immune system that autonomously enacts a multistage response to promote healing. Similar passive mechanisms can enable pneumatic soft robots to overcome common faults such as bursts originating from punctures or overpressurization. Recent technological advancements, ranging from fault-tolerant controllers for robot reconfigurability to self-healing materials, have paved the way for robot resilience. However, these techniques require powerful processors and large datasets or external hardware. How to extend the operational life span of damaged soft robots with minimal computational and physical resources remains unclear. In this study, we demonstrated a multimodal pneumatic soft valve capable of passive resilient reactions, triggered by faults, to prevent or isolate damage in soft robots. In its forward operation mode, the valve, requiring a single supply pressure, isolated punctured soft inflatable elements from the rest of the soft robot in as fast as 21 milliseconds. In its reverse operation mode, the valve can passively protect robots against overpressurization caused by external disturbances, avoiding plastic deformations and bursts. Furthermore, the two modes combined enabled the creation of an endogenously controlled valve capable of autonomous burst isolation. We demonstrated the passive and quick response and the possibility of monolithic integration of the soft valve in grippers and crawling robots. The approach proposed in this study provides a distributed small-footprint alternative to controller-based resilience and is expected to help soft robots achieve uninterrupted long-lasting operation.

Genetic characterization of Tibetan pigs adapted to high altitude under natural selection based on a large whole-genome dataset

The Qinghai-Tibet Plateau is a valuable genetic resource pool, and the high-altitude adaptation of Tibetan pigs is a classic example of the adaptive evolution of domestic animals. Here, we report the presence of Darwinian positive selection signatures in Tibetan pigs (TBPs) using 348 genome-wide datasets (127 whole-genome sequence datasets (WGSs) and 221 whole-genome single-nucleotide polymorphism (SNP) chip datasets). We characterized a high-confidence list of genetic signatures related response to high-altitude adaptation in Tibetan pigs, including 4,598 candidate SNPs and 131 candidate genes. Functional annotation and enrichment analysis revealed that 131 candidate genes are related to multiple systems and organs in Tibetan pigs. Notably, eight of the top ten novel genes, RALB, NBEA, LIFR, CLEC17A, PRIM2, CDH7, GK5 and FAM83B, were highlighted and associated with improved adaptive heart functions in Tibetan pigs high-altitude adaptation. Moreover, genome-wide association analysis revealed that 29 SNPs were involved in 13 candidate genes associated with at least one adaptive trait. In particular, among the top ten candidate genes, CLEC17A is related to a reduction in hemoglobin (HGB) in Tibetan pigs. Overall, our study provides a robust SNP/gene list involving genetic adaptation for Tibetan pig high-altitude adaptation, and it will be a valuable resource for future Tibetan pig studies.

Packaging code and data for reproducible research: A case study of journey time statistics

The effective and ethical use of data to inform decision-making offers huge value to the public sector, especially when delivered by transparent, reproducible, and robust data processing workflows. One way that governments are unlocking this value is through making their data publicly available, allowing more people and organisations to derive insights. However, open data is not enough in many cases: publicly available datasets need to be accessible in an analysis-ready form from popular data science tools, such as R and Python, for them to realise their full potential. This paper explores ways to maximise the impact of open data with reference to a case study of packaging code to facilitate reproducible analysis. We present the jtstats project, which consists of a main Python package, and a smaller R version, for importing, processing, and visualising large and complex datasets representing journey times, for many transport modes and trip purposes at multiple geographic levels, released by the UK Department for Transport (DfT). jtstats shows how domain specific packages can enable reproducible research within the public sector and beyond, saving duplicated effort and reducing the risks of errors from repeated analyses. We hope that the jtstats project inspires others, particularly those in the public sector, to add value to their data sets by making them more accessible.

Anatomy-specific Progression Classification in Chest Radiographs via Weakly Supervised Learning.

"Just Accepted" papers have undergone full peer review and have been accepted for publication in Radiology: Artificial Intelligence. This article will undergo copyediting, layout, and proof review before it is published in its final version. Please note that during production of the final copyedited article, errors may be discovered which could affect the content. Purpose To develop a machine learning approach for classifying disease progression in chest radiographs using weak labels automatically derived from radiology reports. Materials and Methods In this retrospective study, a twin neural network was developed to classify anatomy-specific disease progression into four categories: improved, unchanged, worsened, and new. A two-step weakly-supervised learning approach was employed, pretraining the model on 243,008 frontal chest radiographs from 63,877 MIMIC-CXR patients (mean age 51.7 years; female 34,813), and fine-tuning it on the subset with progression labels derived from consecutive studies. Model performance was evaluated for six pathologic observations on test datasets of unseen MIMIC-CXR patients. Area under the receiver operating characteristic (AUC) analysis was used to evaluate classification performance. The algorithm is also capable of generating bounding-box predictions to localize areas of new progression. Recall, precision, and mean average precision (mAP) were used to evaluate the new progression localization. One-tailed paired t tests were used to assess statistical significance. Results The model outperformed most baselines in progression classification, achieving macro AUC scores of 0.72 ± 0.004 for atelectasis, 0.75 ± 0.007 for consolidation, 0.76 ± 0.017 for edema, 0.81 ± 0.006 for effusion, 0.7 ± 0.032 for pneumonia, and 0.69 ± 0.01 for pneumothorax. For new observation localization, the model achieved mAP scores of 0.25 ± 0.03 for atelectasis, 0.34 ± 0.03 for consolidation, 0.33 ± 0.03 for edema, and 0.31 ± 0.03 for pneumothorax. Conclusion Disease progression classification models were developed on a large chest radiograph dataset, which can be used to monitor interval changes and detect new pathologies on chest radiographs. ©RSNA, 2024.

Personality traits and technical efficiency of Chinese rice farmers

Abstract Given the complex managerial tasks and the risks and uncertainties involved in agricultural production, it is expected that personality traits may play an integral role in shaping farmers’ performance. This study uses a large rural household survey dataset collected in three rice-producing provinces in China to examine this proposition. We find that openness to experience and internal locus of control have favourable effects on technical efficiency (TE), while agreeableness and neuroticism are associated with lower TE. Cognitive skills, as measured by years of education, do not contribute to higher TE of the interviewed farmers. These results imply that the motivational component of human capital (i.e. personality traits) is of greater importance for a farmer’s managerial performance than cognitive skills attained through current formal education. The effectiveness of policies aimed at improving rural welfare may be enhanced by adapting agricultural training programmes to include fostering entrepreneurial and social skills, especially for those with unfavourable personality traits for agricultural production, in addition to improving agricultural knowledge.

Local Evaluation of Large-scale Remote Sensing Machine Learning-generated Building and Road Dataset: The Case of Rwanda

AbstractAccurate and up-to-date building and road data are crucial for informed spatial planning. In developing regions in particular, major challenges arise due to the limited availability of these data, primarily as a result of the inherent inefficiency of traditional field-based surveys and manual data generation methods. Importantly, this limitation has prompted the exploration of alternative solutions, including the use of remote sensing machine learning-generated (RSML) datasets. Within the field of RSML datasets, a plethora of models have been proposed. However, these methods, evaluated in a research setting, may not translate perfectly to massive real-world applications, attributable to potential inaccuracies in unknown geographic spaces. The scepticism surrounding the usefulness of datasets generated by global models, owing to unguaranteed local accuracy, appears to be particularly concerning. As a consequence, rigorous evaluations of these datasets in local scenarios are essential for gaining insights into their usability. To address this concern, this study investigates the local accuracy of large RSML datasets. For this evaluation, we employed a dataset generated using models pre-trained on a variety of samples drawn from across the world and accessible from public repositories of open benchmark datasets. Subsequently, these models were fine-tuned with a limited set of local samples specific to Rwanda. In addition, the evaluation included Microsoft’s and Google’s global datasets. Using ResNet and Mask R‑CNN, we explored the performance variations of different building detection approaches: bottom-up, end-to-end, and their combination. For road extraction, we explored the approach of training multiple models on subsets representing different road types. Our testing dataset was carefully designed to be diverse, incorporating both easy and challenging scenes. It includes areas purposefully chosen for their high level of clutter, making it difficult to detect structures like buildings. This inclusion of complex scenarios alongside simpler ones allows us to thoroughly assess the robustness of DL-based detection models for handling diverse real-world conditions. In addition, buildings were evaluated using a polygon-wise comparison, while roads were assessed using network length-derived metrics.Our results showed a precision (P) of around 75% and a recall (R) of around 60% for the locally fine-tuned building model. This performance was achieved in three out of six testing sites and is considered the lowest limit needed for practical utility of RSML datasets, according to the literature. In contrast, comparable results were obtained in only one out of six sites for the Google and Microsoft datasets. Our locally fine-tuned road model achieved moderate success, meeting the minimum usability threshold in four out of six sites. In contrast, the Microsoft dataset performed well on all sites. In summary, our findings suggest improved performance in road extraction, relative to building extraction tasks. Moreover, we observed that a pipeline relying on a combination of bottom-up and top-down segmentation, while leveraging open global benchmark annotation dataset as well as a small number of samples for fine-tuning, can offer more accurate RSML datasets compared to an open global dataset. Our findings suggest that relying solely on aggregated accuracy metrics can be misleading. According to our evaluation, even city-level derived measures may not capture significant variations in performance within a city, such as lower accuracy in specific neighbourhoods. Overcoming the challenges of complex areas might benefit from exploring alternative approaches, including the integration of LiDAR data, UAV images, aerial images or using other network architectures.

Maximizing the performance of badminton athletes through core strength training: Unlocking their full potential using machine learning (ML) modeling

Core strength training plays an essential role in maximizing performance for badminton athletes. The core muscles in the abdominal, back, and hip regions provide stability, enable efficient power transfer between the upper and lower body, and allow for rapid changes in direction - all crucial components for success in badminton. However, optimizing core training requires an understanding of its impact on sport-specific skills. A variety of exercises targeting the abdominal, back, and hip muscles are discussed. Incorporating core strength training into regular regimens can improve athletes' overall strength, endurance, balance, control, and prevent injuries. This study investigates the effects of various core exercises on stability, agility, and power in badminton players. A comprehensive literature review was conducted to explore the biomechanical demands of badminton and how core musculature contributes to movements like serving, smashing, and lunging. Studies assessing the effects of core training programs in related racquet sports were also examined. The results indicate that targeted core exercises significantly improve athletes' stability, agility, and power output. Exercises targeting the abdominal, back, and hip muscles enhance performance capabilities while reducing injury risk. Machine learning (ML) techniques are then applied to further analyze the relationship between core training and athletic performance. An Artificial Neural Network (ANN) is developed using a dataset of athletes' training histories, metrics, and injury profiles. The model predicts enhancements to stability, agility, and strength from optimized core strengthening routines. Validation confirms the network accurately captures the complex interactions between training variables and physical attributes. This integrated approach provides evidence-based guidelines for tailoring individualized training regimens to unleash players' full abilities. ANNs hold promise for analyzing large datasets on athletes' performance metrics, training variables, and injury histories to design personalized training programs. Linear regression analysis confirmed the ANN's accurate predictions. The findings emphasize integrating data-driven core strength training tailored for badminton into comprehensive programs can help optimize physical abilities and elevate performance levels.

Times series record of water quality analysis during flood time in Cauvery River basin, Tamil Nadu, India

We determine the spatial and temporal changes in the physicochemical and nutritional parameters of water quality in the Cauvery River basin. We also analyze the effects of anthropogenic and natural factors on the physicochemical and nutritional parameters. The river water samples were collected over a time series record of 6-h intervals for 6 days during and after flooding time. The combination of physicochemical and nutritional parameters generated a large dataset, which was processed using statistical analysis. In addition, statistical analysis was used to calculate the correlation coefficients and plot the regression equations relating various parameters to electrical conductivity (EC). The resulting average values of pH, EC, and total dissolved solids (TDS) showed significant differences in the time series record of the study period (i.e., pH = 8.24, EC = 220 S/m, and TDS = 157.1 mg/L). The values of the following nutritional parameters in floodwater were also determined: orthophosphate = 0.13 mg/L, phosphate = 0.1 mg/L, nitrite = 2.11 mg/L, nitrate = 0.79 mg/L, and ammonia = 4.47 mg/L. Our findings indicates that flooding can deteriorate the groundwater quality, resulting in a decrease in water quality and an increase in contamination levels, which can have detrimental effects on both the environment and human health. This information is crucial to decision making in flood risk management, water resource planning, and conservation efforts to protect both human and ecological systems.

Comparison of neural FEM and neural operator methods for applications in solid mechanics

AbstractMachine learning methods are progressively investigated for a large amount of applications. Recently, the solution of partial differential equations (PDE) describing problems in elastostatics came into focus. The current work investigates two neural network-based classes of methods for their solution, namely the neural finite element method (FEM) and neural operator methods. The analysis of these approaches is carried out by means of numerical experiments with linear and nonlinear material behavior where the conventional FEM serves as a benchmark. The formulation of neural FEM allows for elegant integration of finite deformation hyperelasticity at medium training effort. Here, training data are replaced by the evaluation of the equilibrium PDE at sample points. In contrast, most neural operator methods require expensive training with large data sets, but then allow for solving multiple boundary value problems with the same machine learning model. For the comparative analysis, the maximal relative error values over the whole domain and over all components of the strain tensor are evaluated as accuracy measure. The current state of research shows that none of the methods investigated reaches the accuracy and computational performance of the conventional FEM. In many standard applications, the FEM achieves an accuracy of $$10^{-6}$$ 10 - 6 , whereas the numerical tests in the present work report a relative error of order of magnitude of $$10^{-4}$$ 10 - 4 for the neural FEM and $$10^{-2}$$ 10 - 2 to $$10^{-3}$$ 10 - 3 for neural operator methods.

Accelerating photoacoustic microscopy by reconstructing undersampled images using diffusion models

Photoacoustic Microscopy (PAM) integrates optical and acoustic imaging, offering enhanced penetration depth for detecting optical-absorbing components in tissues. Nonetheless, challenges arise in scanning large areas with high spatial resolution. With speed limitations imposed by laser pulse repetition rates, the potential role of computational methods is highlighted in accelerating PAM imaging. We propose a novel and highly adaptable algorithm named DiffPam that utilizes diffusion models to speed up the photoacoustic imaging process. We leveraged a diffusion model trained exclusively on natural images, comparing its performance with an in-domain trained U-Net model using a dataset focused on PAM images of mice brain microvasculature. Our findings indicate that DiffPam performs similarly to a dedicated U-Net model without needing a large dataset. We demonstrate that scanning can be accelerated fivefold with limited information loss. We achieved a 24.70%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$24.70\\%$$\\end{document} increase in peak signal-to-noise ratio and a 27.54%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$27.54\\%$$\\end{document} increase in structural similarity index compared to the baseline bilinear interpolation method. The study also introduces the efficacy of shortened diffusion processes for reducing computing time without compromising accuracy. DiffPam stands out from existing methods as it does not require supervised training or detailed parameter optimization typically needed for other unsupervised methods. This study underscores the significance of DiffPam as a practical algorithm for reconstructing undersampled PAM images, particularly for researchers with limited artificial intelligence expertise and computational resources.

Quantum word embedding for machine learning

The accelerated progress in quantum computing has enabled a new form of machine intelligence that runs on quantum hardware, which holds great promise for more powerful computational models in various learning tasks. An emergent application of Quantum Machine Intelligence (QMI) is Quantum Natural Language Processing (QNLP). This paper proposes a multi-dimensional, finite automaton model for quantum word embedding (QWE) via the Galois field. We demonstrated the model to three applications: (1) English vocabulary, (2) amino acid-based genetic codes, and (3) DNA-based genetic codes. The numerical results obtained from the proposed algorithm for the English vocabulary indicate that it produces more representative word features than Word2Vec based on the word distance metric. Second, the proposed algorithm is also utilized to model RNA-Protein interaction based on the latent distance of a given molecule, which is demonstrated on three large datasets, namely RPI369, RPI1807, and RPI2241. Finally, two embedding techniques for DNA-based genetic codes are proposed in this work: Two-state Lackadaisical Encoding (TCE) and Topological-Cyclic Encoding (TLE). These techniques enable extracting relevant features for the efficacy score of gRNAs used in the CRISPR-Cas 9 system, demonstrated on 15 datasets, compared to 12 mathematical features. We make our implementation available at https://github.com/namnguyen0510/Quantum-Embedding-of-Word/tree/main.

A Characterization of Central Auditory Processing in Parkinson's Disease.

Research indicates that people with Parkinson's disease (PwPs) may experience challenges in both peripheral and central auditory processing, although findings are inconsistent across studies. Due to the diversity of auditory measures used, there is a need for standardized, replicable hearing assessments to clarify which aspects of audition are impacted in PWPs and whether they are linked to motor and non-motor symptoms. To characterize auditory processes and their possible alteration in PwPs. To address this, we collected a comprehensive set of standardized measures of audition using PART, a digital testing platform designed to facilitate replication. Additionally, we examined the relationship between auditory, cognitive, and clinical variables in PwPs. We included 44 PwPs and 54 age and education matched healthy controls. Assessments included detection of diotic and dichotic frequency modulation, temporal gaps, spectro-temporal broad-band modulation, and speech-on-speech masking. We found no statistically significant differences in auditory processing measures between PwPs and the comparison group (ps > 0.07). In PwPs, an auditory processing composite score showed significant medium size correlations with cognitive measures (0.39 < r<0.41, ps < 0.02) and clinical variables of motor symptom severity, quality of life, depression, and caretaker burden (0.33 < r<0.52, ps < 0.03). While larger datasets are needed to clarify whether PwPs experience more auditory difficulties than healthy controls, our results underscore the importance of considering auditory processing on the symptomatic spectrum of Parkinson's disease using standardized replicable methodologies.

Identifying Predictors of Heart Failure Readmission in Patients From a Statutory Health Insurance Database: Retrospective Machine Learning Study.

Patients with heart failure (HF) are the most commonly readmitted group of adult patients in Germany. Most patients with HF are readmitted for noncardiovascular reasons. Understanding the relevance of HF management outside the hospital setting is critical to understanding HF and factors that lead to readmission. Application of machine learning (ML) on data from statutory health insurance (SHI) allows the evaluation of large longitudinal data sets representative of the general population to support clinical decision-making. This study aims to evaluate the ability of ML methods to predict 1-year all-cause and HF-specific readmission after initial HF-related admission of patients with HF in outpatient SHI data and identify important predictors. We identified individuals with HF using outpatient data from 2012 to 2018 from the AOK Baden-Württemberg SHI in Germany. We then trained and applied regression and ML algorithms to predict the first all-cause and HF-specific readmission in the year after the first admission for HF. We fitted a random forest, an elastic net, a stepwise regression, and a logistic regression to predict readmission by using diagnosis codes, drug exposures, demographics (age, sex, nationality, and type of coverage within SHI), degree of rurality for residence, and participation in disease management programs for common chronic conditions (diabetes mellitus type 1 and 2, breast cancer, chronic obstructive pulmonary disease, and coronary heart disease). We then evaluated the predictors of HF readmission according to their importance and direction to predict readmission. Our final data set consisted of 97,529 individuals with HF, and 78,044 (80%) were readmitted within the observation period. Of the tested modeling approaches, the random forest approach best predicted 1-year all-cause and HF-specific readmission with a C-statistic of 0.68 and 0.69, respectively. Important predictors for 1-year all-cause readmission included prescription of pantoprazole, chronic obstructive pulmonary disease, atherosclerosis, sex, rurality, and participation in disease management programs for type 2 diabetes mellitus and coronary heart disease. Relevant features for HF-specific readmission included a large number of canonical HF comorbidities. While many of the predictors we identified were known to be relevant comorbidities for HF, we also uncovered several novel associations. Disease management programs have widely been shown to be effective at managing chronic disease; however, our results indicate that in the short term they may be useful for targeting patients with HF with comorbidity at increased risk of readmission. Our results also show that living in a more rural location increases the risk of readmission. Overall, factors beyond comorbid disease were relevant for risk of HF readmission. This finding may impact how outpatient physicians identify and monitor patients at risk of HF readmission.

Remote sensing for rangeland conservation monitoring: Impacts of livestock removal after 15 years

Abstract Degradation from overgrazing of global arid rangelands is of significant concern for this widespread terrestrial ecosystem, and the people who rely on its sustainable use. Addressing this degradation requires a better understanding of how livestock removal and conservation management actions affect the rehabilitation of land condition. However, the dominant effect of climate and meteorological variability on arid vegetation makes isolating management‐induced change a difficult task. We present an analytically simple method for measuring relative land condition by assessing differences in remote‐sensing‐derived persistent fractional vegetation between adjacent stocked and destocked regions over time. The 22‐year observation period focusses on vegetation recovery at a long‐established sheep station after the removal of livestock and the introduction of conservation‐focussed management. The study region, in the southern Australian arid rangelands, comprises sparsely vegetated chenopod shrublands and acacia woodlands across low‐lying plains. Comparison is made to immediately surrounding stocked properties in a spatial design where climate and biogeographic variables remain consistent across the study area. We found an unequivocal relative increase in vegetation cover, and reduction of bare ground cover at the destocked property that became evident approximately 8 years after a reduction in grazing pressure. Distinctly higher persistent vegetation cover was maintained, even during periods of drought. Non‐photosynthetic vegetation cover was most indicative of long‐term change and increased by 1.1% relative to surrounding areas. Photosynthetic vegetation cover increased by 0.5%, while bare ground cover decreased by 2.1%. These cover changes, particularly in the non‐photosynthetic vegetation and bare ground components spatially coincide with the fenced boundary of the destocked property. Synthesis and applications: The ability to quantify management‐related impacts in land condition using freely available remote sensing data could support improved management by all stakeholders in these regions. The method leverages readily available satellite remote sensing data in a study design that is not dependent on large geospatial and climatic datasets that are not necessarily available in remote regions.

CircHIPK3 nucleates IGF2BP2 and functions as a competing endogenous RNA

Circular RNAs represent a class of endogenous RNAs that regulate gene expression and influence cell biological decisions with implications for the pathogenesis of several diseases. Here, we disclose a novel gene-regulatory role of circHIPK3 by combining analyses of large genomics datasets and mechanistic cell biological follow-up experiments. Using time-course depletion of circHIPK3 and specific candidate RNA-binding proteins, we identify several perturbed genes by RNA sequencing analyses. Expression-coupled motif analyses identify an 11-mer motif within circHIPK3, which also becomes enriched in genes that are downregulated upon circHIPK3 depletion. By mining eCLIP datasets and combined with RNA immunoprecipitation assays, we demonstrate that the 11-mer motif constitutes a strong binding site for IGF2BP2 in bladder cancer cell lines. Our results suggest that circHIPK3 can sequester IGF2BP2 as a competing endogenous RNA (ceRNA), leading to target mRNA stabilization. As an example of a circHIPK3-regulated gene, we focus on the STAT3 mRNA as a specific substrate of IGF2BP2 and validate that manipulation of circHIPK3 regulates IGF2BP2-STAT3 mRNA binding and, thereby, STAT3 mRNA levels. Surprisingly, absolute copy number quantifications demonstrate that IGF2BP2 outnumbers circHIPK3 by orders of magnitude, which is inconsistent with a simple 1:1 ceRNA hypothesis. Instead, we show that circHIPK3 can nucleate multiple copies of IGF2BP2, potentially via phase separation, to produce IGF2BP2 condensates. Our results support a model where a few cellular circHIPK3 molecules can induce IGF2BP2 condensation, thereby regulating key factors for cell proliferation.

Correlation analysis in X-ray photoemission spectroscopy

X-ray photoelectron spectroscopy (XPS) is a powerful technique for surface analysis, but its application can be hindered by uncertainty in modelling spectra. Often, many spectral models have a similar goodness of fit, and distinguishing between them can be impossible without additional information. A further challenge is found in interpreting spectra from samples consisting of multiple chemical compounds. We show here how correlation analysis can be used to interpret large XPS datasets. Correlations in atomic concentrations and binding energies of core lines can be interpreted within a framework of an underlying chemical model and this can yield additional information compared with analysis of each spectrum individually. We give examples of the usage of this analysis on some simple systems, and discuss the potential and limitations of the technique.

CircHIPK3 nucleates IGF2BP2 and functions as a competing endogenous RNA.

Circular RNAs represent a class of endogenous RNAs that regulate gene expression and influence cell biological decisions with implications for the pathogenesis of several diseases. Here, we disclose a novel gene-regulatory role of circHIPK3 by combining analyses of large genomics datasets and mechanistic cell biological follow-up experiments. Using time-course depletion of circHIPK3 and specific candidate RNA-binding proteins, we identify several perturbed genes by RNA sequencing analyses. Expression-coupled motif analyses identify an 11-mer motif within circHIPK3, which also becomes enriched in genes that are downregulated upon circHIPK3 depletion. By mining eCLIP datasets and combined with RNA immunoprecipitation assays, we demonstrate that the 11-mer motif constitutes a strong binding site for IGF2BP2 in bladder cancer cell lines. Our results suggest that circHIPK3 can sequester IGF2BP2 as a competing endogenous RNA (ceRNA), leading to target mRNA stabilization. As an example of a circHIPK3-regulated gene, we focus on the STAT3 mRNA as a specific substrate of IGF2BP2 and validate that manipulation of circHIPK3 regulates IGF2BP2-STAT3 mRNA binding and, thereby, STAT3 mRNA levels. Surprisingly, absolute copy number quantifications demonstrate that IGF2BP2 outnumbers circHIPK3 by orders of magnitude, which is inconsistent with a simple 1:1 ceRNA hypothesis. Instead, we show that circHIPK3 can nucleate multiple copies of IGF2BP2, potentially via phase separation, to produce IGF2BP2 condensates. Our results support a model where a few cellular circHIPK3 molecules can induce IGF2BP2 condensation, thereby regulating key factors for cell proliferation.

Improved quantitative parameter estimation for prostate T2 relaxometry using convolutional neural networks.

Quantitative parameter mapping conventionally relies on curve fitting techniques to estimate parameters from magnetic resonance image series. This study compares conventional curve fitting techniques to methods using neural networks (NN) for measuring T2 in the prostate. Large physics-based synthetic datasets simulating T2 mapping acquisitions were generated for training NNs and for quantitative performance comparisons. Four combinations of different NN architectures and training corpora were implemented and compared with four different curve fitting strategies. All methods were compared quantitatively using synthetic data with known ground truth, and further compared on in vivo test data, with and without noise augmentation, to evaluate feasibility and noise robustness. In the evaluation on synthetic data, a convolutional neural network (CNN), trained in a supervised fashion using synthetic data generated from naturalistic images, showed the highest overall accuracy and precision amongst the methods. On in vivo data, this best performing method produced low-noise T2 maps and showed the least deterioration with increasing input noise levels. This study showed that a CNN, trained with synthetic data in a supervised manner, may provide superior T2 estimation performance compared to conventional curve fitting, especially in low signal-to-noise regions.

Transferable, Living Data Sets for Predicting Global Minimum Energy Nanocluster Geometries.

Modeling of nanocluster geometries is essential for studying the dependence of catalytic activity on the available active sites. In heterogeneous catalysis, the interfacial interaction of the support with the metal can result in modification of the structural and electronic properties of the clusters. To tackle the study of a diverse array of cluster shapes, data-driven methodologies are essential to circumvent prohibitive computational costs. At their core, these methods require large data sets in order to achieve the necessary accuracy to drive structural exploration. Given the similarity in binding character of the transition metals, cluster shapes encountered for various systems show a large amount of overlap. This overlap has been utilized to construct a living data set which may be carried over across multiple studies. Iterative refinement of this data set provides a low-cost pathway for initialization of cluster studies. It is shown that utilization of transferable structural information can reduce model construction costs by more than 90%. The benefits of this approach are particularly notable for alloy systems, which possess significantly larger configurational spaces compared to the pure-phase counterparts.