Large Datasets Research Articles

Discarded sequencing reads uncover natural variation in pest resistance in Thlaspi arvense.

Understanding the genomic basis of natural variation in plant pest resistance is an important goal in plant science, but it usually requires large and labor-intensive phenotyping experiments. Here, we explored the possibility that non-target reads from plant DNA sequencing can serve as phenotyping proxies for addressing such questions. We used data from a whole-genome and -epigenome sequencing study of 207 natural lines of field pennycress (Thlaspi arvense) that were grown in a common environment and spontaneously colonized by aphids, mildew, and other microbes. We found that the numbers of non-target reads assigned to the pest species differed between populations, had significant SNP-based heritability, and were associated with climate of origin and baseline glucosinolate contents. Specifically, pennycress lines from cold and thermally fluctuating habitats, presumably less favorable to aphids, showed higher aphid DNA load, i.e., decreased aphid resistance. Genome-wide association analyses identified genetic variants at known defense genes but also novel genomic regions associated with variation in aphid and mildew DNA load. Moreover, we found several differentially methylated regions associated with pathogen loads, in particular differential methylation at transposons and hypomethylation in the promoter of a gene involved in stomatal closure, likely induced by pathogens. Our study provides first insights into the defense mechanisms of Thlaspi arvense, a rising crop and model species, and demonstrates that non-target whole-genome sequencing reads, usually discarded, can be leveraged to estimate intensities of plant biotic interactions. With rapidly increasing numbers of large sequencing datasets worldwide, this approach should have broad application in fundamental and applied research.

Stratified Medicine Paediatrics: Cell free DNA and serial tumour sequencing identifies subtype specific cancer evolution and epigenetic states.

We profiled a large heterogenous cohort of matched diagnostic-relapse tumour tissue and paired plasma-derived cell free DNA (cfDNA) from patients with relapsed and progressive solid tumours of childhood. Tissue and cfDNA sequencing results were concordant, with a wider spectrum of mutant alleles and higher degree of intra-tumour heterogeneity captured by the latter, if sufficient circulating tumour-derived DNA (ctDNA) was present. Serial tumour sequencing identified putative drivers of relapse, with alterations in epigenetic drivers being a common feature. In keeping with epigenetic alterations being a common driver of many childhood cancers, fragmentomics analysis of cfDNA identified tumour-specific epigenetic states and transcription factor binding sites accessible in chromatin. This study leverages a large and well-annotated genomic dataset of aggressive childhood malignancies, identifies genomic and epigenetic drivers of childhood cancer relapse, and highlights the power and practicality of cfDNA analysis to capture both intra-tumoural heterogeneity and the epigenetic state of cancer cells.

Racial preferences in sports NFTs? Empirical evidence from NBA Top Shots

Purpose The purpose of this paper is to analyse a large data set of over 2.4 million Top Shot NFT transactions to determine whether there is evidence of revealed racial preferences in the Top Shot market.Design/methodology/approach This study examines a data set of 2.43 million transactions from 28 July 2020 to 20 March 2021 using Ordinary Least Squares, Weighted Least Squares and Inverse Probability Weighting. Our models include indicator variables for the race of depicted players and examine estimated race coefficients for statistical significance and effect size.Findings Our results indicate that the National Basketball Association (NBA) Top Shot NFTs depicting Black players or Black/Mixed players trade for higher prices than those depicting White players, after controlling for other factors including on-court performance. While highly statistically significant, the effect sizes are below conventional thresholds for small effects.Practical implicationsRecent empirical studies present evidence suggesting that Black players may earn less than comparable White players. Our findings challenge implicit justifications for observed earning disparities based on consumer preference.Originality/value This study provides empirical analysis into revealed racial preference of consumers for sports memorabilia. It is the first to examine racial preference in NBA NFTs.

GARD-LENS: A downscaled large ensemble dataset for understanding future climate and its uncertainties

This article introduces the Generalized Analog Regression Downscaling method Large Ensemble (GARD-LENS) dataset, comprised of daily precipitation, mean temperature, and temperature range over the Contiguous U.S., Alaska, and Hawaii at 12-km, 4-km, and 1-km resolutions, respectively. GARD-LENS statistically downscales three CMIP6 global climate model large ensembles, CESM2, CanESM5, and EC-Earth3, totaling 200 ensemble members. GARD-LENS is the first downscaled SMILE (single model initial-condition large ensemble), providing information about the role of internal climate variability at high resolutions. The 150-year record of this large ensemble dataset provides ample data for assessing trends and extremes and allows users to robustly assess internal variability, forced climate signals, and time of emergence at high resolutions. As the need for high resolution, robust climate datasets continues to grow, GARD-LENS will be a valuable tool for scientists and practitioners who wish to account for internal variability in their future climate analyses and adaptation plans.

Variability in hot sub-luminous stars and binaries: Machine-learning analysis of Gaia DR3 multi-epoch photometry

Hot sub-luminous stars represent a population of stripped and evolved red giants that is located on the extreme horizontal branch. Since they exhibit a wide range of variability due to pulsations or binary interactions, it is crucial to unveil their intrinsic and extrinsic variability to understand the physical processes of their formation. In the Hertzsprung-Russell diagram, they overlap with interacting binaries such as cataclysmic variables (CVs). By leveraging the most recent clustering algorithm tools, we investigate the variability of 1,576 candidate hot subdwarf variables using comprehensive data from Gaia DR3 multi-epoch photometry and Transiting Exoplanet Survey Satellite (TESS) observations. We present a novel approach that uses the t-distributed stochastic neighbour embedding and the uniform manifold approximation and projection dimensionality reduction algorithms to facilitate the identification and classification of different populations of variable hot subdwarfs and CVs in a large dataset. In addition to the publicly available Gaia time-series statistics table, we adopted additional statistical features that enhanced the performance of the algorithms. The clustering results led to the identification of 85 new hot subdwarf variables based on Gaia and TESS light curves and of 108 new variables based on Gaia light curves alone, including reflection-effect systems, HW Vir, ellipsoidal variables, and high-amplitude pulsating variables. A significant number of known CVs (140) distinctively cluster in the 2D feature space among an additional 152 objects that we consider candidates for new CVs. This study paves the way for more efficient and comprehensive analyses of stellar variability from ground- and space-based observations, and for the application of machine-learning classifications of candidate variable stars in large surveys.

Automated galaxy sizes in Euclid images using the Segment Anything Model

Stellar disk truncations, also referred to as galaxy edges, are key indicators of galactic size, determined by the radial location of the gas density threshold for star formation. This threshold essentially marks the boundary of the luminous matter in a galaxy. Accurately measuring galaxy sizes for millions of galaxies is essential for understanding the physical processes driving galaxy evolution over cosmic time. We aim to explore the potential of the Segment Anything Model (SAM), a foundation model designed for image segmentation, to automatically identify disk truncations in galaxy images. With the Euclid Wide Survey poised to deliver vast datasets, our goal is to assess SAM’s capability to measure galaxy sizes in a fully automated manner. SAM was applied to a labeled dataset of 1,047 disk-like galaxies with $M_* > 10^ at redshifts up to $z 1$, sourced from the Hubble Space Telescope (HST) CANDELS fields. We “euclidized” the HST galaxy images by creating composite RGB images, using the F160W (H-band), F125W (J-band), and F814W + F606W (I-band + V-band) HST filters, respectively. Using these processed images as input for SAM, we retrieved various truncation masks for each galaxy image under different configurations of the input data. We find excellent agreement between the galaxy sizes identified by SAM and those measured manually (i.e., by using the radial positions of the stellar disk edges in galaxy light profiles), with an average deviation of approximately $3<!PCT!>$. This error reduces to about $1<!PCT!>$ when excluding problematic cases. Our results highlight the strong potential of SAM for detecting disk truncations and measuring galaxy sizes across large datasets in an automated way. SAM performs well without requiring extensive image preprocessing, labeled training datasets for truncations (used only for validation), fine-tuning, or additional domain-specific adaptations such as transfer learning.

Insect Pollinator Voucher Collection: Pollinator Visitation to Pacific Northwest Native Plants and Native Cultivars

Planting native flora continues to grow in popularity as a means of conserving pollinator fauna in fragmented landscapes. Due to the limited availability of native plants, consumers may encounter cultivars in their search for plants native to their region. Though consumers have a documented interest in planting native flora for their purported benefits to pollinators, it is unclear whether cultivars provide the same benefits as wild-type native plants. Over three years (2020-2022), we observed and collected pollinators from a common garden experiment in Corvallis, OR containing 8 species of Pacific Northwest native plants, 18 cultivars derived from native plant species, and one exotic perennial plant. Methods for collection are described in further detail in Hayes et al. (in prep.). Here, we report on and document the deposition of taxonomic voucher specimens which represent the primary species concepts used throughout Hayes et al. (in prep.) (Accession #). Specimens were identified to the lowest possible taxonomic level by L. R. Best (bees, Hymenoptera: Apoidea) and J. J-M. Hayes (syrphid flies, Diptera: Syrphidae). The vouchered material includes 139 individual specimens with at least one representative per caste and sex, as available, of 55 species and 14 morphospecies of bees (Hymenoptera: Apoidea) and 13 species and one morphospecies of syrphid flies (Diptera: Syrphidae). Additional vouchers are included for specimens identified only to the morphospecies level, and Megachile brevis Say, 1837 individuals that were captured while collecting sections of petal from Clarkia amoena (Lehm.) A.Nelson & J.F.Macbr. plants. These 139 voucher specimens come from a much larger dataset, containing over 6,500 observational records of plant-pollinator interactions.

An intelligent predictive maintenance system based on random forest for addressing industrial conveyor belt challenges

This study introduces a sophisticated intelligent predictive maintenance system for industrial conveyor belts powered by a random forest machine learning model. The random forest model was evaluated against established models such as logistic regression, neural networks, decision trees, and gradient boosting, demonstrating superior performance. The model achieved 100% accuracy in classifying gearbox lubricant levels and sprocket conditions, highlighting its potential for addressing critical challenges in predictive maintenance, such as avoiding unexpected downtime. However, further validation with larger datasets and varied operational environments is recommended to confirm robustness. This performance highlights its effectiveness in multiclass fault detection and overfitting mitigation, establishing a new standard in predictive maintenance technology. The system, enhanced by a comprehensive sensor array, not only adeptly captures but also intelligently analyzes critical operational data, providing proactive and data-driven insights for maintenance decision-making. This study not only affirms the dominance of the random forest model in predictive analytics but also underscores its pivotal role in optimizing maintenance strategies, enhancing operational efficiency, and ensuring the reliability of conveyor systems in industrial settings.

Insect Pollinator Voucher Collection: Pollinator Visitation to Pacific Northwest Native Plants and Native Cultivars

Planting native flora continues to grow in popularity as a means of conserving pollinator fauna in fragmented landscapes. Due to the limited availability of native plants, consumers may encounter cultivars in their search for plants native to their region. Though consumers have a documented interest in planting native flora for their purported benefits to pollinators, it is unclear whether cultivars provide the same benefits as wild-type native plants. Over three years (2020-2022), we observed and collected pollinators from a common garden experiment in Corvallis, OR containing 8 species of Pacific Northwest native plants, 18 cultivars derived from native plant species, and one exotic perennial plant. Methods for collection are described in further detail in Hayes et al. (in prep.). Here, we report on and document the deposition of taxonomic voucher specimens which represent the primary species concepts used throughout Hayes et al. (in prep.) (Accession #). Specimens were identified to the lowest possible taxonomic level by L. R. Best (bees, Hymenoptera: Apoidea) and J. J-M. Hayes (syrphid flies, Diptera: Syrphidae). The vouchered material includes 139 individual specimens with at least one representative per caste and sex, as available, of 55 species and 14 morphospecies of bees (Hymenoptera: Apoidea) and 13 species and one morphospecies of syrphid flies (Diptera: Syrphidae). Additional vouchers are included for specimens identified only to the morphospecies level, and Megachile brevis Say, 1837 individuals that were captured while collecting sections of petal from Clarkia amoena (Lehm.) A.Nelson & J.F.Macbr. plants. These 139 voucher specimens come from a much larger dataset, containing over 6,500 observational records of plant-pollinator interactions.

Estimation of minimal data sets sizes for machine learning predictions in digital mental health interventions

Artificial intelligence promises to revolutionize mental health care, but small dataset sizes and lack of robust methods raise concerns about result generalizability. To provide insights on minimal necessary data set sizes, we explore domain-specific learning curves for digital intervention dropout predictions based on 3654 users from a single study (ISRCTN13716228, 26/02/2016). Prediction performance is analyzed based on dataset size (N = 100–3654), feature groups (F = 2–129), and algorithm choice (from Naive Bayes to Neural Networks). The results substantiate the concern that small datasets (N ≤ 300) overestimate predictive power. For uninformative feature groups, in-sample prediction performance was negatively correlated with dataset size. Sophisticated models overfitted in small datasets but maximized holdout test results in larger datasets. While N = 500 mitigated overfitting, performance did not converge until N = 750–1500. Consequently, we propose minimum dataset sizes of N = 500–1000. As such, this study offers an empirical reference for researchers designing or interpreting AI studies on Digital Mental Health Intervention data.

Reconstructing decision-making dynamics during public health crises by applyingdata science to public records.

Public health policies can dramatically shape government responses to emerging public health crises. In cases where a response is inadequate, it's natural to seek improvements to these policies to achieve better results in future crises. However, while policies and outcomes are usually visible, the complex dynamics that link them are seldom obvious. Obscure social network structures, power and information asymmetries, and political agendas all influence the translation of policy to action or inaction. Intergovernmental communications are often the only primary sources for researchers investigating these hidden but crucial factors. Public records laws in many countries make such documents available upon request, but their structure, format, and scale are rarely accessible in practice. Fortunately, recent advances in free and open-source data science tools are making this problem tractable. In this visual brief, we demonstrate using such tools to mine a large image dataset and reconstruct decision-making during the Flint Water Crisis in Michigan, USA.

CGRclust: Chaos Game Representation for twin contrastive clustering of unlabelled DNA sequences

BackgroundTraditional supervised learning methods applied to DNA sequence taxonomic classification rely on the labor-intensive and time-consuming step of labelling the primary DNA sequences. Additionally, standard DNA classification/clustering methods involve time-intensive multiple sequence alignments, which impacts their applicability to large genomic datasets or distantly related organisms. These limitations indicate a need for robust, efficient, and scalable unsupervised DNA sequence clustering methods that do not depend on sequence labels or alignment.ResultsThis study proposes CGRclust, a novel combination of unsupervised twin contrastive clustering of Chaos Game Representations (CGR) of DNA sequences, with convolutional neural networks (CNNs). To the best of our knowledge, CGRclust is the first method to use unsupervised learning for image classification (herein applied to two-dimensional CGR images) for clustering datasets of DNA sequences. CGRclust overcomes the limitations of traditional sequence classification methods by leveraging unsupervised twin contrastive learning to detect distinctive sequence patterns, without requiring DNA sequence alignment or biological/taxonomic labels. CGRclust accurately clustered twenty-five diverse datasets, with sequence lengths ranging from 664 bp to 100 kbp, including mitochondrial genomes of fish, fungi, and protists, as well as viral whole genome assemblies and synthetic DNA sequences. Compared with three recent clustering methods for DNA sequences (DeLUCS, iDeLUCS, and MeShClust v3.0.), CGRclust is the only method that surpasses 81.70% accuracy across all four taxonomic levels tested for mitochondrial DNA genomes of fish. Moreover, CGRclust also consistently demonstrates superior performance across all the viral genomic datasets. The high clustering accuracy of CGRclust on these twenty-five datasets, which vary significantly in terms of sequence length, number of genomes, number of clusters, and level of taxonomy, demonstrates its robustness, scalability, and versatility.ConclusionCGRclust is a novel, scalable, alignment-free DNA sequence clustering method that uses CGR images of DNA sequences and CNNs for twin contrastive clustering of unlabelled primary DNA sequences, achieving superior or comparable accuracy and performance over current approaches. CGRclust demonstrated enhanced reliability, by consistently achieving over 80% accuracy in more than 90% of the datasets analyzed. In particular, CGRclust performed especially well in clustering viral DNA datasets, where it consistently outperformed all competing methods.

DNEA: an R package for fast and versatile data-driven network analysis of metabolomics data

BackgroundMetabolomics is a high-throughput technology that measures small molecule metabolites in cells, tissues or biofluids. Analysis of metabolomics data is a multi-step process that involves data processing, quality control and normalization, followed by statistical and bioinformatics analysis. The latter step often involves pathway analysis to aid biological interpretation of the data. This approach is limited to endogenous metabolites that can be readily mapped to metabolic pathways. An alternative to pathway analysis that can be used for any classes of metabolites, including unknown compounds that are ubiquitous in untargeted metabolomics data, involves defining metabolite-metabolite interactions using experimental data. Our group has developed several network-based methods that use partial correlations of experimentally determined metabolite measurements. These were implemented in CorrelationCalculator and Filigree, two software tools for the analysis of metabolomics data we developed previously. The latter tool implements the Differential Network Enrichment Analysis (DNEA) algorithm. This analysis is useful for building differential networks from metabolomics data containing two experimental groups and identifying differentially enriched metabolic modules. While Filigree is a user-friendly tool, it has certain limitations when used for the analysis of large-scale metabolomics datasets.ResultsWe developed the DNEA R package for the data-driven network analysis of metabolomics data. We present the DNEA workflow and functionality, algorithm enhancements implemented with respect to the package’s predecessor, Filigree, and discuss best practices for analyses. We tested the performance of the DNEA R package and illustrated its features using publicly available metabolomics data from the environmental determinants of diabetes in the young. To our knowledge, this package is the only publicly available tool designed for the construction of biological networks and subsequent enrichment testing for datasets containing exogenous, secondary, and unknown compounds. This greatly expands the scope of traditional enrichment analysis tools that can be used to analyze a relatively small set of well-annotated metabolites.ConclusionsThe DNEA R package is a more flexible and powerful implementation of our previously published software tool, Filigree. The modular structure of the package, along with the parallel processing framework built into the most computationally extensive steps of the algorithm, make it a powerful tool for the analysis of large and complex metabolomics datasets.

MinLinMo: a minimalist approach to variable selection and linear model prediction

Generating prediction models from high dimensional data often result in large models with many predictors. Causal inference for such models can therefore be difficult or even impossible in practice. The stand-alone software package MinLinMo emphasizes small linear prediction models over highest possible predictability with a particular focus on including variables correlated with the outcome, minimal memory usage and speed. MinLinMo is demonstrated on large epigenetic datasets with prediction models for chronological age, gestational age, and birth weight comprising, respectively, 15, 14 and 10 predictors. The parsimonious MinLinMo models perform comparably to established prediction models requiring hundreds of predictors.

The impact of high-risk cytogenetic abnormalities in extramedullary multiple myeloma in the era of novel agents: insights from a multicenter study

PurposeThis study aimed to examine the impact of high-risk cytogenetic abnormalities (HRA) on the survival outcomes of multiple myeloma patients with extramedullary disease (EMD) in the era of novel agents, utilizing the largest dataset of extramedullary multiple myeloma patients in China.MethodsThis study included a total of 371 patients with EMD, comprising 113 patients with de novo EME and 258 patients with EMB.ResultsPatients with one HRA and those with ≥ 2 HRA demonstrated significantly worse overall survival (OS) (P < 0.01) and progression-free survival (PFS) (P < 0.01) compared to patients without HRA. Additionally, 1q21 gain/amplification (1q21 +) remained a predictor of poor prognosis in EMD. CD38 monoclonal antibody-based therapy and single transplantation were less effective in improving survival outcomes for EMD with ≥ 2 HRA. Multivariable analysis identified LDH levels > 250 U/L, creatinine levels > 177 μmol/L, extramedullary extraosseous (EME), 1 HRA, and ≥ 2 HRA as independent adverse prognostic factors in patients with EMD.ConclusionPatients with EMD who had ≥ 2 HRA experienced an extremely poor prognosis, which could not be improved by single transplantation or CD38 monoclonal antibody-based treatment. The number of HRA could serve as an important factor in guiding treatment choices and predicting prognosis in patients with EMD. Furthermore, 1q21 + remained a significant factor associated with worse survival outcomes in EMD.

Novel neural network classification of maternal fetal ultrasound planes through optimized feature selection

Ultrasound (US) imaging is an essential diagnostic technique in prenatal care, enabling enhanced surveillance of fetal growth and development. Fetal ultrasonography standard planes are crucial for evaluating fetal development parameters and detecting abnormalities. Real-time imaging, low cost, non-invasiveness, and accessibility make US imaging indispensable in clinical practice. However, acquiring fetal US planes with correct fetal anatomical features is a difficult and time-consuming task, even for experienced sonographers. Medical imaging using AI shows promise for addressing current challenges. In response to this challenge, a Deep Learning (DL)-based automated categorization method for maternal fetal US planes are introduced to enhance detection efficiency and diagnosis accuracy. This paper presents a hybrid optimization technique for feature selection and introduces a novel Radial Basis Function Neural Network (RBFNN) for reliable maternal fetal US plane classification. A large dataset of maternal–fetal screening US images was collected from publicly available sources and categorized into six groups: the four fetal anatomical planes, the mother's cervix, and an additional category. Feature extraction is performed using Gray-Level Co-occurrence Matrix (GLCM), and optimization methods such as Particle Swarm Optimization (PSO), Grey Wolf Optimization (GWO), and a hybrid Particle Swarm Optimization and Grey Wolf Optimization (PSOGWO) approach are utilized to select the most relevant features. The optimized features from each algorithm are then input into both conventional and proposed DL models. Experimental results indicate that the proposed approach surpasses conventional DL models in performance. Furthermore, the proposed model is evaluated against previously published models, showcasing its superior classification accuracy. In conclusion, our proposed approach provides a solid foundation for automating the classification of fetal US planes, leveraging optimization and DL techniques to enhance prenatal diagnosis and care.

Detection and classification of long terminal repeat sequences in plant LTR-retrotransposons and their analysis using explainable machine learning

BackgroundLong terminal repeats (LTRs) represent important parts of LTR retrotransposons and retroviruses found in high copy numbers in a majority of eukaryotic genomes. LTRs contain regulatory sequences essential for the life cycle of the retrotransposon. Previous experimental and sequence studies have provided only limited information about LTR structure and composition, mostly from model systems. To enhance our understanding of these key sequence modules, we focused on the contrasts between LTRs of various retrotransposon families and other genomic regions. Furthermore, this approach can be utilized for the classification and prediction of LTRs.ResultsWe used machine learning methods suitable for DNA sequence classification and applied them to a large dataset of plant LTR retrotransposon sequences. We trained three machine learning models using (i) traditional model ensembles (Gradient Boosting), (ii) hybrid convolutional/long and short memory network models, and (iii) a DNA pre-trained transformer-based model using k-mer sequence representation. All three approaches were successful in classifying and isolating LTRs in this data, as well as providing valuable insights into LTR sequence composition. The best classification (expressed as F1 score) achieved for LTR detection was 0.85 using the hybrid network model. The most accurate classification task was superfamily classification (F1=0.89) while the least accurate was family classification (F1=0.74). The trained models were subjected to explainability analysis. Positional analysis identified a mixture of interesting features, many of which had a preferred absolute position within the LTR and/or were biologically relevant, such as a centrally positioned TATA-box regulatory sequence, and TG..CA nucleotide patterns around both LTR edges.ConclusionsOur results show that the models used here recognized biologically relevant motifs, such as core promoter elements in the LTR detection task, and a development and stress-related subclass of transcription factor binding sites in the family classification task. Explainability analysis also highlighted the importance of 5’- and 3’- edges in LTR identity and revealed need to analyze more than just dinucleotides at these ends. Our work shows the applicability of machine learning models to regulatory sequence analysis and classification, and demonstrates the important role of the identified motifs in LTR detection.

DeepMiRBP: a hybrid model for predicting microRNA-protein interactions based on transfer learning and cosine similarity

BackgroundInteractions between microRNAs and RNA-binding proteins are crucial for microRNA-mediated gene regulation and sorting. Despite their significance, the molecular mechanisms governing these interactions remain underexplored, apart from sequence motifs identified on microRNAs. To date, only a limited number of microRNA-binding proteins have been confirmed, typically through labor-intensive experimental procedures. Advanced bioinformatics tools are urgently needed to facilitate this research.MethodsWe present DeepMiRBP, a novel hybrid deep learning model specifically designed to predict microRNA-binding proteins by modeling molecular interactions. This innovation approach is the first to target the direct interactions between small RNAs and proteins. DeepMiRBP consists of two main components. The first component employs bidirectional long short-term memory (Bi-LSTM) neural networks to capture sequential dependencies and context within RNA sequences, attention mechanisms to enhance the model’s focus on the most relevant features and transfer learning to apply knowledge gained from a large dataset of RNA-protein binding sites to the specific task of predicting microRNA-protein interactions. Cosine similarity is applied to assess RNA similarities. The second component utilizes Convolutional Neural Networks (CNNs) to process the spatial data inherent in protein structures based on Position-Specific Scoring Matrices (PSSM) and contact maps to generate detailed and accurate representations of potential microRNA-binding sites and assess protein similarities.ResultsDeepMiRBP achieved a prediction accuracy of 87.4% during training and 85.4% using testing, with an F score of 0.860. Additionally, we validated our method using three case studies, focusing on microRNAs such as miR-451, -19b, -23a, -21, -223, and -let-7d. DeepMiRBP successfully predicted known miRNA interactions with recently discovered RNA-binding proteins, including AGO, YBX1, and FXR2, identified in various exosomes.ConclusionsOur proposed DeepMiRBP strategy represents the first of its kind designed for microRNA-protein interaction prediction. Its promising performance underscores the model’s potential to uncover novel interactions critical for small RNA sorting and packaging, as well as to infer new RNA transporter proteins. The methodologies and insights from DeepMiRBP offer a scalable template for future small RNA research, from mechanistic discovery to modeling disease-related cell-to-cell communication, emphasizing its adaptability and potential for developing novel small RNA-centric therapeutic interventions and personalized medicine.

FlightScope: An Experimental Comparative Review of Aircraft Detection Algorithms in Satellite Imagery

Object detection in remotely sensed satellite pictures is fundamental in many fields such as biophysical and environmental monitoring. While deep learning algorithms are constantly evolving, they have been mostly implemented and tested on popular ground-taken photos. This paper critically evaluates and compares a suite of advanced object detection algorithms customized for the task of identifying aircraft within satellite imagery. The goal is to enable researchers to choose efficiently from algorithms that are trainable and usable in real time on a deep learning infrastructure with moderate requirements. Using the large HRPlanesV2 dataset, together with rigorous validation with the GDIT dataset, this research encompasses an array of methodologies including YOLO versions 5, 8, and 10, Faster RCNN, CenterNet, RetinaNet, RTMDet, DETR, and grounding DINO, all trained from scratch. This exhaustive training and validation study reveals YOLOv5 as the pre-eminent model for the specific case of identifying airplanes from remote sensing data, showcasing high precision and adaptability across diverse imaging conditions. This research highlights the nuanced performance landscapes of these algorithms, with YOLOv5 emerging as a robust solution for aerial object detection, underlining its importance through superior mean average precision, recall, and intersection over union scores. The findings described here underscore the fundamental role of algorithm selection aligned with the specific demands of satellite imagery analysis and extend a comprehensive framework to evaluate model efficacy. This aims to foster exploration and innovation in the realm of remote sensing object detection, paving the way for improved satellite imagery applications.

Use Self-Training Random Forest for Predicting Winter Wheat Yield

The effectiveness of supervised ML heavily depends on having a large, accurate, and diverse annotated dataset, which poses a challenge in applying ML for yield prediction. To address this issue, we developed a self-training random forest algorithm capable of automatically expanding the annotated dataset. Specifically, we trained a random forest regressor model using a small amount of annotated data. This model was then utilized to generate new annotations, thereby automatically extending the training dataset through self-training. Our experiments involved collecting data from over 30 winter wheat varieties during the 2019–2020 and 2021–2022 growing seasons. The testing results indicated that our model achieved an R2 of 0.84, RMSE of 627.94 kg/ha, and MAE of 516.94 kg/ha in the test dataset, while the validation dataset yielded an R2 of 0.81, RMSE of 692.96 kg/ha, and MAE of 550.62 kg/ha. In comparison, the standard random forest resulted in an R2 of 0.81, RMSE of 681.02 kg/ha, and MAE of 568.97 kg/ha in the test dataset, with validation results of an R2 of 0.79, RMSE of 736.24 kg/ha, and MAE of 585.85 kg/ha. Overall, these results demonstrate that our self-training random forest algorithm is a practical and effective solution for expanding annotated datasets, thereby enhancing the prediction accuracy of ML models in winter wheat yield forecasting.