Full Dataset Research Articles

Theranostic Nanoparticle Uptake in Patient Brain Tumors as Quantified by MP2RAGE T1 Mapping

Modern radiation therapy techniques provide effective treatments for solid tumors, but there remain challenges with delivering high doses to elusive tumors without causing toxicity to surrounding normal tissue. Pre-clinical trials have demonstrated the theranostic properties of a recently developed gadolinium-based nanoparticle (Gd-NP). The first in-human clinical trial was conducted to assess the safety and dose tolerance of intravenous Gd-NPs in combination with whole brain radiotherapy and showed favorable results, including a significant correlation between tumor uptake and therapeutic response. The second, double-blinded, in-human clinical trial is underway in the US and aims to evaluate if brain-directed stereotactic radiation in conjunction with NPs will improve local tumor control compared to radiation alone. The current work investigates uptake patterns in brain tumors of 23 patients as quantified by magnetization prepared 2 rapid gradient echo (MP2RAGE) T1 mapping. A phantom containing eight vials of NP-saline solutions at varying concentrations was created to examine the relationship between NP concentration and longitudinal relaxation (T1, in seconds). This relationship is known as relaxivity and is dependent on the contrast agent, field strength and T1 mapping sequence. A 3T MAGNETOM Vida scanner and MP2RAGE sequence were used to image the phantom and MP2RAGE T1 maps were calculated using Bloch equations (QMRLab software). Relaxivity was determined and applied to 23 patient T1 maps (pre- and post- Gd-NP administration) to calculate uptake on an individual tumor basis. Theranostic NP uptake was calculated for every voxel in each of 129 brain metastases and examined for patterns in quantity and distribution. Average individual tumor uptake ranged from 0.02-0.12 mg/ml, where average overall uptake was equal to 0.05 mg/ml. A relationship between tumor diameter and mean NP concentration was observed and best represented by a power-based curve (R2 = 0.92). In contrast, patients with suspected placebo administration appeared to have no uptake and therefore no relationship with tumor diameter. The distribution of NP concentration within the tumor was also examined; on average, linear uptake profiles through tumor centroids (ant-post, left-right) demonstrated roughly gaussian patterns of uptake with lower concentrations at the tumor edges and higher concentrations at the tumor center. This pattern indicates robust tumor penetration and may have implications for amplifying radiation dose to hypoxic tumors. Gd-NP uptake in brain metastases can be quantified using MP2RAGE T1 mapping. Uptake was determined for each voxel in each tumor volume, where a gaussian pattern of spatial concentrations was observed. This analysis procedure will be applied to the full data set, when available, to evaluate the impact of NP uptake (in conjunction with radiation therapy) within individual patients and individual tumors.

Subsurface geological and geophysical data from the Po Plain and the northern Adriatic Sea (north Italy)

Abstract. The Po Plain (Italy) is one of the most densely populated and productive regions of Europe, characterized by a flourishing economy (also linked to strategic subsurface resources) and several world cultural and natural heritage sites. The coupling of socio-economic interests with geological hazards (i.e. seismic, subsidence, and flooding hazards) in this area requires accurate knowledge of the subsurface geology, the active geological processes, and the impact of human activities on natural environments to mitigate the potential natural and anthropic risks. Most data unveiling the subsurface geology of this region were produced by the hydrocarbon exploration industry. Indeed, the Po Plain hosts many hydrocarbon fields that have been discovered since the early 1950s, giving rise to the subsurface exploration through extensive seismic reflection surveys and drilling of numerous deep wells. In this work, geological and geophysical data from 160 deep wells drilled for hydrocarbon exploration and/or exploitation purposes in the Po Plain and in the facing northern Adriatic Sea have been collected and digitized along with several published geological cross-sections and maps. These data have been used to reconstruct the overall subsurface 3D architecture and to extract the physical properties of the subsurface geological units. The digitized data are suitable to be imported into geo-software environments so as to derive the geophysical and mechanical properties of the geological units for a wealth of applied and scientific studies such as geomechanical, geophysical, and seismological studies. The integrated dataset may represent a useful tool in defining regional first-order strategies to ensure the safety of the urbanized areas and human activities and to reduce natural and anthropic risks that may affect this crucial region of Europe. In particular, the data collected would be useful to highlight sensible areas where data collection and more detailed studies are needed. Nowadays, such issues are particularly relevant for the underground industry development related to the increasing interest in possible CO2 and hydrogen underground storage, which can play a fundamental role in the energy transition process towards decarbonization goals. The full dataset is available at the following link: https://doi.org/10.5281/zenodo.8126519 (Livani et al., 2023).

Machine Learning for the Prediction of Survival Post-Allogeneic Hematopoietic Cell Transplantation: A Single-Center Experience

Introduction: Prediction of outcomes following allogeneic hematopoietic cell transplantation (HCT) remains a major challenge. Machine learning (ML) is a computational procedure that may facilitate the generation of HCT prediction models. We sought to investigate the prognostic potential of multiple ML algorithms when applied to a large single-center allogeneic HCT database. Methods: Our registry included 2,697 patients that underwent allogeneic HCT from January 1976 to December 2017. 45 pretransplant baseline variables were included in the predictive assessment of each ML algorithm on overall survival (OS) as determined by area under the curve (AUC). Pretransplant variables used in the EBMT ML study (Shouval et al., 2015) were used as a benchmark for comparison. Results: On the entire dataset, the random forest (RF) algorithm performed best (AUC 0.71 ± 0.04) compared to the second-best model, logistic regression (LR) (AUC = 0.69 ± 0.04) (p < 0.001). Both algorithms demonstrated improved AUC scores using all 45 variables compared to the limited variables examined by the EBMT study. Survival at 100 days post-HCT using RF on the full dataset discriminated patients into different prognostic groups with different 2-year OS (p < 0.0001). We then examined the ML methods that allow for significant individual variable identification, including LR and RF, and identified matched related donors (HR = 0.49, p < 0.0001), increasing TBI dose (HR = 1.60, p = 0.006), increasing recipient age (HR = 1.92, p < 0.0001), higher baseline Hb (HR = 0.59, p = 0.0002), and increased baseline FEV1 (HR = 0.73, p = 0.02), among others. Conclusion: The application of multiple ML techniques on single-center allogeneic HCT databases warrants further investigation and may provide a useful tool to identify variables with prognostic potential.

Effects of Deep Vertical Rotary Tillage Management Methods on Soil Quality in Saline Cotton Fields in Southern Xinjiang

A long-term high-saline soil environment will limit the improvement of soil quality and cotton yield. Modified tillage management measures can improve soil quality, and the establishment of a soil quality evaluation system will facilitate evaluation of the soil quality and land production potential in southern Xinjiang. The objective of this study was to determine the effects of different tillage management methods on soil quality in saline cotton fields in southern Xinjiang. A three-year experiment was conducted in Tumushuke, Xinjiang, with different deep vertical rotary tillage depths (DTM20, 20 cm; DTM40, 40 cm; DTM60, 60 cm) and conventional tillage (CTM, 20 cm). The soil quality index (SQI) under different tillage management methods was established by using the full dataset (TDS) with a scoring function for eight indicators, including physicochemical properties of the soil from 0 to 60 cm, to evaluate its impact on the soil quality of the saline farmland in southern Xinjiang. The results of the study showed that deep vertical rotary tillage management can effectively optimize soil structure; reduce soil bulk density (BD), soil solution conductivity (EC), and pH; and promote the accumulation of soil organic carbon (SOC) and total nitrogen (TN) in the soil. However, the average diameter of soil water-stable aggregates (MWD) in a 0–60 cm layer becomes smaller with an increasing depth of tillage. This does not reduce crop yields but does promote soil saline leaching. In addition, the significant linear relationship (p &lt; 0.001) between seed cotton yield and soil quality indicated that improving soil quality was favorable for crop yield. The principal component analysis revealed BD, MWD, pH, and EC as limiting sensitive indicators for seed cotton yield, while SOC and TN were positive sensitive indicators. The soil quality index (SQI) values of DT40 and DTM60 were significantly higher than that of CTM by 11.02% and 15.27%, respectively. Overall, the results show that DTM60 is the most suitable tillage strategy to improve soil quality and seed cotton yield in this area, and this approach will provide a reliable theoretical basis for the improvement of saline farmland.

Can monetary poverty measurement detect multidimensionally deprived? Evidence from Egypt

PurposeThis paper aims to examine the mismatch between multidimensional deprivation and monetary poverty in identifying the poor in Egypt and investigates their determinants empirically.Design/methodology/approachThe paper uses the Alkire-Foster multidimensional poverty measurement method using data from Egypt’s 2017/2018 Household Income, Expenditure and Consumption Survey (HIECS 2017/2018). Using a logistic regression model, the paper assesses the empirical relationship between multidimensional and monetary poverty and their determinants at the aggregate level and by dimension.FindingsThe paper demonstrates a significant mismatch between multidimensional and monetary poverty measures, underscoring their complementary nature. Statistics indicate that both measures overlap in classifying 35.81% of Egyptians, whereas monetary poverty ignores 63.12% of multidimensionally poor in at least one dimension. Regression estimates show a significant moderate negative association between expenditure per capita and multidimensional poverty and its dimensions. Moreover, they show that household head’s gender, age, education attainment, marital status, job proficiency, household size and location affect poverty mismatch and match in Egypt.Practical implicationsThis paper offers Egyptian policymakers the multidimensional poverty index that enables more efficient designing and targeting of poverty alleviation programs and assessing current poverty alleviation programs to modify them if needed.Originality/valueTo the best of the authors’ knowledge, this study is the first to examine the mismatch between both poverty measures in Egypt, using the recent full data set of HIECS 2017/2018. This paper confirms that depending only on monetary measures can send inaccurate insights for crafting effective social policies. Also, it offers policymakers a comprehensive insight into the country’s poverty landscape, which enable more efficient design, targeting of poverty alleviation programs and monitoring their effectiveness.

On the dynamics of a dark sector coupling

Interacting dark energy models may play a crucial role in explaining several important observational issues in modern cosmology and also may provide a solution to current cosmological tensions. Since the phenomenology of the dark sector could be extremely rich, one should not restrict the interacting models to have a coupling parameter which is constant in cosmic time, rather allow for its dynamical behaviour, as it is common practice in the literature when dealing with other dark energy properties, as the dark energy equation of state. We present here a compendium of the current cosmological constraints on a large variety of interacting models, investigating scenarios where the coupling parameter of the interaction function and the dark energy equation of state can be either constant or dynamical. For the most general schemes, in which both the coupling parameter of the interaction function and the dark energy equation of state are dynamical, we find 95% CL evidence for a dark energy component at early times and slightly milder evidence for a dynamical dark coupling for the most complete observational data set exploited here, which includes CMB, BAO and Supernova Ia measurements. Interestingly, there are some cases where a dark energy component different from the cosmological constant case at early times together with a coupling different from zero today, can alleviate both the H0 and S8 tension for the full dataset combination considered here. Due to the energy exchange among the dark sectors, the current values of the matter energy density and of the clustering parameter σ8 are shifted from their ΛCDM-like values. This fact makes future surveys, especially those focused on weak lensing measurements, unique tools to test the nature and the couplings of the dark energy sector.

Spatiotemporal variability in pH and carbonate parameters on the Canadian Atlantic continental shelf between 2014 and 2022

Abstract. The Atlantic Zone Monitoring Program (AZMP) was established by Fisheries and Oceans Canada (DFO) in 1998 with the aim of monitoring physical and biological ocean conditions in Atlantic Canada in support of fisheries management. Since 2014, at least two of the carbonate parameters (pH; total alkalinity, TA; and dissolved inorganic carbon, DIC) have also been systematically measured as part of the AZMP, enabling the calculation of derived parameters (e.g., carbonate saturation states, Ω, and partial pressure of CO2, pCO2). The present study gives an overview of the spatiotemporal variability in these parameters between 2014 and 2022. Results show that the variability in the carbonate system reflects changes in both physical (e.g., temperature and salinity) and biological (e.g., plankton photosynthesis and respiration) parameters. For example, most of the region undergoes a seasonal warming and freshening. While the former will tend to increase Ω, the latter will decrease both TA and Ω. Spring and summer plankton blooms decrease DIC near the surface and then remineralize and increase DIC at depth in the fall. The lowest pCO2 values (down to ∼ 200 µatm) are located in the cold coastal Labrador Current, whereas the highest values (&gt;1500 µatm) are found in the fresh waters of the Gulf of St. Lawrence and the St. Lawrence Estuary. The latter is also host to the lowest pH values of the zone (7.48 in the fall of 2022). Finally, most of the bottom waters of the Gulf of St. Lawrence (&gt;90 %) are undersaturated with respect to aragonite (Ωarg&lt;1). In addition to providing a baseline of carbonate parameters for the Atlantic Zone as a whole, this comprehensive overview is a necessary and useful contribution for the modelling community and for more in-depth studies. The full dataset of measured and derived parameters is available from the Federated Research Data Repository: https://doi.org/10.20383/102.0673 (Cyr et al., 2022a).

KPCA-WPHM-SCNs-based remaining useful life prediction method for motor rolling bearings

Motor rolling bearings are the important supporting components of motors. It can ensure the stable operation of motor equipment in the power grid, and bearing life prediction of it is a key issue. To solve the problem of low accuracy of remaining useful life (RUL) prediction for motor rolling bearings, a neural network model based on Weibull proportional hazards model (WPHM) and stochastic configuration networks (SCNs) is proposed. To better extract and analyze features of the bearing vibration signal in both time and frequency domains, kernel principal component analysis (KPCA) is used to reduce the dimensionality of the data. Then, a WPHM model using the top three contributing feature parameters is built, which sets the start time based on the failure rate curve and reliability function. Finally, the validity of the model is verified with the rolling bearing full life cycle dataset from the IEEE PHM 2012 Data Challenge, and a comparison with other machine learning models shows that the accuracy of the proposed model in RUL prediction is higher.

Using Different Types of Artificial Neural Networks to Classify 2D Matrix Codes and Their Rotations-A Comparative Study.

Artificial neural networks can solve various tasks in computer vision, such as image classification, object detection, and general recognition. Our comparative study deals with four types of artificial neural networks-multilayer perceptrons, probabilistic neural networks, radial basis function neural networks, and convolutional neural networks-and investigates their ability to classify 2D matrix codes (Data Matrix codes, QR codes, and Aztec codes) as well as their rotation. The paper presents the basic building blocks of these artificial neural networks and their architecture and compares the classification accuracy of 2D matrix codes under different configurations of these neural networks. A dataset of 3000 synthetic code samples was used to train and test the neural networks. When the neural networks were trained on the full dataset, the convolutional neural network showed its superiority, followed by the RBF neural network and the multilayer perceptron.

New atomic data for C i Rydberg states compared with solar UV spectra

ABSTRACT We use the Breit–Pauli R-matrix method to calculate accurate energies and radiative data for states in C i up to n = 30 and with l ≤ 3. We provide the full data set of decays to the five 2s2 2p2 ground configuration states 3P0,1,2, 1D2, and 1S0. This is the first complete set of data for transitions from n ≥ 5. We compare oscillator strengths and transition probabilities with the few previously calculated values for such transitions, finding generally good agreement (within 10 per cent) with the exception of values recently recommended by National Institute of Standards and Technology, where significant discrepancies are found. We then calculate spectral line intensities originating from the Rydberg states using typical chromospheric conditions and assuming local thermal equilibrium, and compare them with well-calibrated Solar and Heliospheric Observatory Solar Ultraviolet Measurements of Emitted Radiation ultraviolet (UV) spectra of the quiet Sun. The relative intensities of the Rydberg series are in excellent agreement with observation, which provides firm evidence for the identifications and blends of nearly 200 UV lines. Such comparison also resulted in a large number of new identifications of C i lines in the spectra. We also estimate optical depth effects and find that these can account for much of the absorption noted in the observations. The atomic data can be applied to model a wide range of solar and astrophysical observations.

Using temporal recalibration to improve the calibration of risk prediction models in competing risk settings when there are trends in survival over time.

We have previously proposed temporal recalibration to account for trends in survival over time to improve the calibration of predictions from prognostic models for new patients. This involves first estimating the predictor effects using data from all individuals (full dataset) and then re-estimating the baseline using a subset of the most recent data whilst constraining the predictor effects to remain the same. In this article, we demonstrate how temporal recalibration can be applied in competing risk settings by recalibrating each cause-specific (or subdistribution) hazard model separately. We illustrate this using an example of colon cancer survival with data from the Surveillance Epidemiology and End Results (SEER) program. Data from patients diagnosed in 1995-2004 were used to fit two models for deaths due to colon cancer and other causes respectively. We discuss considerations that need to be made in order to apply temporal recalibration such as the choice of data used in the recalibration step. We also demonstrate how to assess the calibration of these models in new data for patients diagnosed subsequently in 2005. Comparison was made to a standard analysis (when improvements over time are not taken into account) and a period analysis which is similar to temporal recalibration but differs in the data used to estimate the predictor effects. The 10-year calibration plots demonstrated that using the standard approach over-estimated the risk of death due to colon cancer and the total risk of death and that calibration was improved using temporal recalibration or period analysis.

SensorSCAN: Self-supervised learning and deep clustering for fault diagnosis in chemical processes

Modern industrial facilities generate large volumes of raw sensor data during the production process. This data is used to monitor and control the processes and can be analyzed to detect and predict process abnormalities. Typically, the data has to be annotated by experts in order to be used in predictive modeling. However, manual annotation of large amounts of data can be difficult in industrial settings.In this paper, we propose SensorSCAN, a novel method for unsupervised fault detection and diagnosis, designed for industrial chemical process monitoring. We demonstrate our model's performance on two publicly available datasets of the Tennessee Eastman Process with various faults. The results show that our method significantly outperforms existing approaches (+0.2-0.3 TPR for a fixed FPR) and effectively detects most of the process faults without expert annotation. Moreover, we show that the model fine-tuned on a small fraction of labeled data nearly reaches the performance of a SOTA model trained on the full dataset. We also demonstrate that our method is suitable for real-world applications where the number of faults is not known in advance. The code is available at https://github.com/AIRI-Institute/sensorscan.

Bootstrapping outperforms community‐weighted approaches for estimating the shapes of phenotypic distributions

Abstract Estimating phenotypic distributions of populations and communities is central to many questions in ecology and evolution. These distributions can be characterized by their moments (mean, variance, skewness and kurtosis) or diversity metrics (e.g. functional richness). Typically, such moments and metrics are calculated using community‐weighted approaches (e.g. abundance‐weighted mean). We propose an alternative bootstrapping approach that allows flexibility in trait sampling and explicit incorporation of intraspecific variation, and show that this approach significantly improves estimation while allowing us to quantify uncertainty. We assess the performance of different approaches for estimating the moments of trait distributions across various sampling scenarios, taxa and datasets by comparing estimates derived from simulated samples with the true values calculated from full datasets. Simulations differ in sampling intensity (individuals per species), sampling biases (abundance, size), trait data source (local vs. global) and estimation method (two types of community‐weighting, two types of bootstrapping). We introduce the traitstrap R package, which contains a modular and extensible set of bootstrapping and weighted‐averaging functions that use community composition and trait data to estimate the moments of community trait distributions with their uncertainty. Importantly, the first function in the workflow, trait_fill, allows the user to specify hierarchical structures (e.g. plot within site, experiment vs. control, species within genus) to assign trait values to each taxon in each community sample. Across all taxa, simulations and metrics, bootstrapping approaches were more accurate and less biased than community‐weighted approaches. With bootstrapping, a sample size of 9 or more measurements per species per trait generally included the true mean within the 95% CI. It reduced average percent errors by 26%–74% relative to community‐weighting. Random sampling across all species outperformed both size‐ and abundance‐biased sampling. Our results suggest randomly sampling ~9 individuals per sampling unit and species, covering all species in the community and analysing the data using nonparametric bootstrapping generally enable reliable inference on trait distributions, including the central moments, of communities. By providing better estimates of community trait distributions, bootstrapping approaches can improve our ability to link traits to both the processes that generate them and their effects on ecosystems.

Demonstration of an AI-driven workflow for autonomous high-resolution scanning microscopy

Modern scanning microscopes can image materials with up to sub-atomic spatial and sub-picosecond time resolutions, but these capabilities come with large volumes of data, which can be difficult to store and analyze. We report the Fast Autonomous Scanning Toolkit (FAST) that addresses this challenge by combining a neural network, route optimization, and efficient hardware controls to enable a self-driving experiment that actively identifies and measures a sparse but representative data subset in lieu of the full dataset. FAST requires no prior information about the sample, is computationally efficient, and uses generic hardware controls with minimal experiment-specific wrapping. We test FAST in simulations and a dark-field X-ray microscopy experiment of a WSe2 film. Our studies show that a FAST scan of <25% is sufficient to accurately image and analyze the sample. FAST is easy to adapt for any scanning microscope; its broad adoption will empower general multi-level studies of materials evolution with respect to time, temperature, or other parameters.

MeshCLIP: Efficient cross-modal information processing for 3D mesh data in zero/few-shot learning

Text, 2D, and 3D information are crucial information representations in modern science and management disciplines. However, complex and irregular 3D data produce data scarcity and expensive generation that limit their processing and application. In this paper, we present MeshCLIP, a new cross-modal information learning paradigm to directly process 3D mesh data end-to-end in a zero/few-shot manner. Specifically, we design a novel pipeline based on visual factors and graphics principles, bridging the gap between 3D mesh data and other modal data, thereby joining 2/3D visual and textual information for zero/few-shot learning. Then, we construct a self-attention adapter for 3D mesh key information learning when training only a few priors, significantly improving the model’s discriminative ability. Extensive experiments demonstrate that the proposed MeshCLIP can achieve state-of-the-art results on multiple challenging 3D mesh datasets. In the whole 3D domain, the proposed zero-shot approach significantly outperforms the existing other 3D representation methods with an accuracy 3 × better ( increased by 41.5%) on the ModelNet40 dataset. Furthermore, in few-shot learning, the proposed MeshCLIP uses only a few supervised priors (only less than 10% of the sample size) to achieve results close to those of methods trained on a full dataset.

Video-Restoration-Net: Deep Generative Model with Non-Local Network for Inpainting and Super-Resolution Tasks

Although deep learning-based approaches for video processing have been extensively investigated, the lack of generality in network construction makes it challenging for practical applications, particularly in video restoration. As a result, this paper presents a universal video restoration model that can simultaneously tackle video inpainting and super-resolution tasks. The network, called Video-Restoration-Net (VRN), consists of four components: (1) an encoder to extract features from each frame, (2) a non-local network that recombines features from adjacent frames or different locations of a given frame, (3) a decoder to restore the coarse video from the output of a non-local block, and (4) a refinement network to refine the coarse video on the frame level. The framework is trained in a three-step pipeline to improve training stability for both tasks. Specifically, we first suggest an automated technique to generate full video datasets for super-resolution reconstruction and another complete-incomplete video dataset for inpainting, respectively. A VRN is then trained to inpaint the incomplete videos. Meanwhile, the full video datasets are adopted to train another VRN frame-wisely and validate it against authoritative datasets. We show quantitative comparisons with several baseline models, achieving 40.5042 dB/0.99473 on PSNR/SSIM in the inpainting task, while during the SR task we obtained 28.41 dB/0.7953 and 27.25/0.8152 on BSD100 and Urban100, respectively. The qualitative comparisons demonstrate that our proposed model is able to complete masked regions and implement super-resolution reconstruction in videos of high quality. Furthermore, the above results show that our method has greater versatility both in video inpainting and super-resolution tasks compared to recent models.

Performance analysis and comparison of data-driven models for predicting indoor temperature in multi-zone commercial buildings

Building thermal models, which characterize the properties of a building’s envelope and thermal mass, are essential for accurate indoor temperature and cooling/heating demand prediction. Because of their flexibility and ease of use, data-driven models are increasingly used. This study compared and analyzed the performance of gray-box (resistance-capacitance) and black-box (recurrent neural network) models for predicting indoor air temperature in a real multi-zone commercial building. The developed resistance-capacitance model served as a benchmark model for which full sets of temporal data and building information were used as inputs. The recurrent neural network models were trained and tested assuming various available types and amounts of temporal data and known building physical information to investigate the effects of data and information availability. Feature importance analysis was conducted to select the key variables for different prediction targets under different scenarios. This research provides guidance in selecting an appropriate building thermal response modeling method based on the measured data availability, building physical information, and application.

Genomic prediction of seasonal forage yield in perennial ryegrass

AbstractBackgroundGenomic selection has the potential to accelerate genetic gain in perennial ryegrass breeding, provided complex traits such as forage yield can be predicted with sufficient accuracy.MethodsIn this study, we compared modelling approaches and feature selection strategies to evaluate the accuracy of genomic prediction models for seasonal forage yield production.ResultsOverall, model selection had limited impact on predictive ability when using the full data set. For a baseline genomic best linear unbiased prediction model, the highest mean predictive accuracy was obtained for spring grazing (0.78), summer grazing (0.62) and second cut silage (0.56). In terms of feature selection strategies, using uncorrelated single‐nucleotide polymorphisms (SNPs) had no impact on predictive ability, allowing for a potential decrease of the data set dimensions. With a genome‐wide association study, we found a significant SNP marker for spring grazing, located in the genic region annotated as coding for an enzyme responsible for fucosylation of xyloglucans—major components of the plant cell wall. We also presented an approach to increase interpretability of genomic prediction models with the use of Gene Ontology enrichment analysis.ConclusionsApproaches for feature selection will be relevant in development of low‐cost genotyping platforms in support of routine and cost‐effective implementation of genomic selection.

DeepTPpred: A Deep Learning Approach With Matrix Factorization for Predicting Therapeutic Peptides by Integrating Length Information.

The abuse of traditional antibiotics has led to increased resistance of bacteria and viruses. Efficient therapeutic peptide prediction is critical for peptide drug discovery. However, most of the existing methods only make effective predictions for one class of therapeutic peptides. It is worth noting that currently no predictive method considers sequence length information as a distinct feature of therapeutic peptides. In this article, a novel deep learning approach with matrix factorization for predicting therapeutic peptides (DeepTPpred) by integrating length information are proposed. The matrix factorization layer can learn the potential features of the encoded sequence through the mechanism of first compression and then restoration. And the length features of the sequence of therapeutic peptides are embedded with encoded amino acid sequences. To automatically learn therapeutic peptide predictions, these latent features are input into the neural networks with self-attention mechanism. On eight therapeutic peptide datasets, DeepTPpred achieved excellent prediction results. Based on these datasets, we first integrated eight datasets to obtain a full therapeutic peptide integration dataset. Then, we obtained two functional integration datasets based on the functional similarity of the peptides. Finally, we also conduct experiments on the latest versions of the ACP and CPP datasets. Overall, the experimental results show that our work is effective for the identification of therapeutic peptides.

SPATIO-TEMPORAL VARIATION OF AEOLIAN SHOREWARD SAND TRANSPORT MEASURED USING NEAR-CONTINUOUS LASER SCANNING

A beach in Mariakerke-Bad (Belgium) was monitored in 2017-2018 for more than a year with a near-continuous laser scan system. From a total of 8500 scans 7700 hourly scan epochs were used to study the spatio-temporal shoreward sand transport at the beach. In order to account for weather influences and other possible disturbances of the scan-system, a time-dependent correction method was applied to reduce rotation errors up to 0.2 degrees in the point cloud orientation (around the zero point of the laser scanner) reducing height errors on the beach to the order of centimeters. Cross shore analysis of the beach profile shows that shoreward transport occurs at most times during the year with an accumulated maximum of 17m3/m throughout the measurement period and maximum transport rates of 0.6 m3/m/day. However most of the shoreward sand transport is redistributed seawards again due to beach shaping leaving a total of about 2 m3/m a year which is below the average values found along the Belgium coast. The spatio-temporal behavior of the shoreward sand transport has been studied with the 4D-OBC analysis technique which identified accumulations of sand in the full 4D point cloud dataset. A total of about 3600 4D-OBC accumulation events were identified and most found accumulations on the beach can be associated with natural (aeolian) processes. Also, accumulations appear to occur during the whole year which is consistent with the previous cross-shore analysis.