DINO-EYE: self-supervised learning for identification of different optic disc phenotypes in primary open angle glaucoma

We aimed to establish a rapid and precise method for identifying and quantifying multi-species vegetable oil (corn oil, olive oil (OLO), soybean oil, and sunflower oil (SUO)) adulterations in camellia oil (CAO), using soft ionization by chemical reaction in transfer–high-resolution mass spectrometry (SICRIT-HRMS) and machine learning methods. The results showed that SICRIT-HRMS could effectively characterize the volatile profiles of pure and adulterated CAO samples, including binary, ternary, quaternary, and quinary adulteration systems. The low m/z region (especially 100–300) exhibited importance to oil classification in multiple feature-selection methods. For qualitative detection, binary classification models based on convolutional neural networks (CNN), Random Forest (RF), and gradient boosting trees (GBT) algorithms showed high accuracies (98.70–100.00%) for identifying CAO adulteration under no dimensionality reduction (NON), principal component analysis (PCA), and uniform manifold approximation and projection (UMAP) strategies. The RF algorithm exhibited relatively high accuracy (96.25–99.45%) in multiclass classification. Moreover, the five models, including CNN, RF, support vector machines (SVM), logistic regression (LR), and GBT, exhibited different performances in distinguishing pure and adulterated CAO. Among 1093 blind oil samples, under NON, PCA, and UMAP: 10, 5, and 67 samples were misclassified by CNN model; 6, 7, and 41 samples were misclassified by RF model; 8, 9, and 82 samples were misclassified by SVM model; 17, 18, and 78 samples were misclassified by LR model; 7, 9, and 43 samples were misclassified by GBT model. For quantitative prediction, the PCA-CNN model performed optimally in predicting adulteration levels in CAO, especially with respect to OLO and SUO, exhibiting a high coefficient of determination for calibration (, 0.9664–0.9974) and coefficient of determination for prediction (, 0.9599–0.9963) values, low root mean square error of calibration (RMSEC, 0.9–5.3%) and root mean square error of prediction (RMSEP, 1.1–5.8%) values, and RPD (5.0–16.3) values greater than 3.0. These results indicate that SICRIT-HRMS combined with machine learning can rapidly and accurately identify and quantify multi-species vegetable oil adulterations in CAO, which provides a reference for developing non-targeted and high-throughput detection methods in edible oil authenticity.

Network Modeling of Apoptosis-Related Genes Reveal a High-Risk Cluster in Multiple Myeloma

Crustacean aquaculture is expected to be a major source of fishery commodities in the near future. Hemocytes are key players of the immune system in shrimps; however, their classification, maturation, and differentiation are still under debate. To date, only discrete and inconsistent information on the classification of shrimp hemocytes has been reported, showing that the morphological characteristics are not sufficient to resolve their actual roles. Our present study using single-cell RNA sequencing revealed six types of hemocytes of Marsupenaeus japonicus based on their transcriptional profiles. We identified markers of each subpopulation and predicted the differentiation pathways involved in their maturation. We also predicted cell growth factors that might play crucial roles in hemocyte differentiation. Different immune roles among these subpopulations were suggested from the analysis of differentially expressed immune-related genes. These results provide a unified classification of shrimp hemocytes, which improves the understanding of its immune system.

<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">As a common seismic facies visualization analysis method, SOM projects the waveform or seismic attribute vectors into a two-dimensional topological plane in a nonlinear way, which can effectively and efficiently discover the topological structure of a dataset. SOM does not need to set the number of classes in prior, and has friendly visualization characteristics and excellent generalization, which are conducive to seismic facies interpretation using unlabeled data. However, due to the competitive learning used in SOM and the imbalance of data distribution in real world, the samples from majority classes are expanded on the topological plane and the minority classes are compressed. As a result, the plane cannot accurately describe the global structure of data distribution. To improve the visualization precision by modifying the topological relationship of the prototype vectors of SOM, we utilize UMAP (Uniform Manifold Approximation and Projection), a novel manifold learning technique for dimension reduction, to correct the prototype vectors generated by SOM. By combing the advantages of SOM and UMAP in the representation of data topological structure, the global relationship between seismic data samples can be properly established, and the internal relative spatial structure of majority class samples can be retained as much as possible, resulting in a more reliable classification. Meanwhile, the framework maintains the advantages of SOM in visualization. In the modeling tests and real data experiments, we have demonstrated the effectiveness and rationality of UMAP-SOM on the spatial structure representation of three-dimensional seismic data</i> .

To overcome the challenges of high signal-to-noise ratio requirements and the poor applicability of pulse waveform features in partial discharge (PD) blind source separation for cable accessories, this paper proposes a pulse source separation technique based on Uniform Manifold Approximation and Projection (UMAP). This method enables pulse signal separation in the presence of various noise interferences by using the time-frequency spectrogram of PD pulses as input and applying the UMAP algorithm for dimensionality reduction and feature extraction of the spectrogram's global characteristics. Unlike traditional feature extraction methods, this approach eliminates the need for defining and selecting specific features. The reduced-dimensional features effectively capture the relative differences between different discharge pulses, enabling their separation. Experimental validation using data from typical cable accessory defects shows that this method can effectively extract and separate features from random pulses. Compared to similar manifold approximation algorithms, UMAP is less sensitive to hyperparameters, and the reduced-dimensional features are more stable, which facilitates the execution of clustering algorithms.

Numerical modelling increasingly generates massive, high-dimensional spatio-temporal datasets. Exploring such datasets relies on effective visualization. This study presents a generic workflow to (i) project high-dimensional spatio-temporal data on a two-dimensional (2D) plane accurately (ii) compare dimensionality reduction techniques (DRTs) in terms of resolution and computational efficiency (iii) represent 2D projection spatially using a 2D perceptually uniform background color map. Machine learning (ML) based DRTs for data visualization i.e., principal component analysis (PCA), generative topographic mapping (GTM), t-distributed stochastic neighbor embedding (t-SNE) and uniform manifold approximation and projection (UMAP) are compared in terms of accuracy, resolution and computational efficiency to handle massive datasets. The accuracy of visualization is evaluated using a quality metric based on a co-ranking framework. The workflow is applied to an output of an Australian Water Resource Assessment (AWRA) model for Tasmania, Australia. The dataset consists of daily time series of nine components of the water balance at a 5 km grid cell resolution for the year 2017. The case study shows that PCA allows rapid visualization of global data structures, while t-SNE and UMAP allows more accurate representation of local trends. Furthermore, UMAP is computationally more efficient than t-SNE and least affected by the outliers compared to GTM.

Learning from potentially infinite and high-dimensional data streams poses significant challenges in the classification task. For instance, k-Nearest Neighbors (kNN) is one of the most often used algorithms in the data stream mining area that proved to be very resource-intensive when dealing with high-dimensional spaces. Uniform Manifold Approximation and Projection (UMAP) is a novel manifold technique and one of the most promising dimension reduction and visualization techniques in the non-streaming setting because of its high performance in comparison with competitors. However, there is no version of UMAP that copes with the challenging context of streams. To overcome these restrictions, we propose a batch-incremental approach that pre-processes data streams using UMAP, by producing successive embeddings on a stream of disjoint batches in order to support an incremental kNN classification. Experiments conducted on publicly available synthetic and real-world datasets demonstrate the substantial gains that can be achieved with our proposal compared to state-of-the-art techniques.

Identifying collective variables (CVs) for chemical reactions is essential to reduce the 3N-dimensional energy landscape into lower dimensional basins and barriers of interest. However, in condensed phase processes, the nonmeaningful motions of bulk solvent often overpower the ability of dimensionality reduction methods to identify correlated motions that underpin collective variables. Yet solvent can play important indirect or direct roles in reactivity, and much can be lost through treatments that remove or dampen solvent motion. This has been amply demonstrated within principal component analysis (PCA), although less is known about the behavior of nonlinear dimensionality reduction methods, e.g., uniform manifold approximation and projection (UMAP), that have become recently utilized. The latter presents an interesting alternative to linear methods though often at the expense of interpretability. This work presents distance-attenuated projection methods of atomic coordinates that facilitate the application of both PCA and UMAP to identify collective variables in the presence of explicit solvent and further the specific identity of solvent molecules that participate in chemical reactions. The performance of both methods is examined in detail for two reactions where the explicit solvent plays very different roles within the collective variables. When applied to raw molecular dynamics data in solution, both PCA and UMAP representations are dominated by bulk solvent motions. On the other hand, when applied to data preprocessed by our attenuated projection methods, both PCA and UMAP identify the appropriate collective variables (though varying sensitivity is observed due to the presence of explicit solvent that results from the projection method). Importantly, this approach allows identification of specific solvent molecules that are relevant to the CVs and their importance.

This study used an odor sensing system with a 16-channel electrochemical sensor array to measure beef odors, aiming to distinguish odors under different storage days and processing temperatures for quality monitoring. Six storage days ranged from purchase (D0) to eight days (D8), with three temperature conditions: no heat (RT), boiling (100 °C), and frying (180 °C). Gas chromatography–mass spectrometry (GC-MS) analysis showed that odorants in the beef varied under different conditions. Compounds like acetoin and 1-hexanol changed significantly with the storage days, while pyrazines and furans were more detectable at higher temperatures. The odor sensing system data were visualized using principal component analysis (PCA) and uniform manifold approximation and projection (UMAP). PCA and unsupervised UMAP clustered beef odors by storage days but struggled with the processing temperatures. Supervised UMAP accurately clustered different temperatures and dates. Machine learning analysis using six classifiers, including support vector machine, achieved 57% accuracy for PCA-reduced data, while unsupervised UMAP reached 49.1% accuracy. Supervised UMAP significantly enhanced the classification accuracy, achieving over 99.5% with the dimensionality reduced to three or above. Results suggest that the odor sensing system can sufficiently enhance non-destructive beef quality and safety monitoring. This research advances electronic nose applications and explores data downscaling techniques, providing valuable insights for future studies.

Advances in single-cell technologies have enabled high-resolution dissection of tissue composition. Several tools for dimensionality reduction are available to analyze the large number of parameters generated in single-cell studies. Recently, a nonlinear dimensionality-reduction technique, uniform manifold approximation and projection (UMAP), was developed for the analysis of any type of high-dimensional data. Here we apply it to biological data, using three well-characterized mass cytometry and single-cell RNA sequencing datasets. Comparing the performance of UMAP with five other tools, we find that UMAP provides the fastest run times, highest reproducibility and the most meaningful organization of cell clusters. The work highlights the use of UMAP for improved visualization and interpretation of single-cell data.

Visualization is widely used to help explore and interpret high dimensional single cell (sc) omics data, such as scRNA-seq expression data. In particular, uniform manifold approximation and projection (UMAP) has become nearly ubiquitous in scientific publications that apply single cell omics technologies. Some experts have expressed concerns that the global cell-cell relationship, especially the spatial distances among cell clusters in a dataset, may not be faithfully depicted in a 2-dimensional (2D) UMAP. To help users to better appreciate this issue with their own data, we created an online server for the community to upload their single cell data and interactively make UMAPs with different hyper-parameters to witness how the distribution of cell clusters changes. The server thus can help promote proper usages of UMAP, especially to avoid the common pitfalls in misinterpretation of inter-cluster relationships in single cell studies.Availability and ImplementationShinyUMAP is freely available as an online Shiny server implemented in Python athttps://scviewer.shinyapps.io/shinyUMAP/.

Machine learning techniques have become more attractive and widely used for medical image processing purposes. In particular, the diagnosis of neurodegenerative diseases has recently shown a potential field of application for these methods. The performance comparison of a unique algorithm in various study contexts can be biased, which usually leads to incorrect results. In this context, this study consists in comparing the performance of different machine learning techniques, identifying their main trends and their application for the diagnosis of Alzheimer’s disease (AD). We presented a computer-aided diagnosis system for the early diagnosis of AD by analyzing brain data from the OASIS dataset. The principal component analysis (PCA) and the uniform manifold approximation and projection (UMAP) technique have been evaluated on the magnetic resonance imaging and positron emission tomography images as feature selection techniques. After that, the features are fed into nine machine learning models namely Support vector machine (SVM), Artificial neural networks, Decision trees, Random Forests, Discriminant analysis, Regression analysis, Naive Bayes, k-Nearest neighbors, and Ensemble learning. The performance of the proposed classifiers is investigated by the confusion matrix. In addition, area under the curve, Matthews correlation coefficient, accuracy, and F1-score metrics are calculated regarding this matrix. Our results indicate that the SVM-PCA/UMAP schemes provide a significant advantage over the other classifiers. Moreover, they are more efficient than the baseline model based on the voxels-as-features reference feature extraction approach.KeywordsAlzheimer’s diseaseNeuroimaging dataComputer-aided diagnosis systemPrincipal component analysisUniform manifold approximation and projection techniqueMachine learning classifiers

Many researchers have realized the intelligent medical diagnosis of diabetic retinopathy (DR) from fundus images by using deep learning methods, including supervised contrastive learning (SupCon). However, although SupCon brings label information into the calculation of contrastive learning, it does not distinguish between augmented positives and same-label positives. As a result, we propose the concept of Angular Margin and incorporate it into SupCon to address this issue. To demonstrate the effectiveness of our strategy, we tested it on two datasets for the detection and grading of DR. To align with previous work, Accuracy, Precision, Recall, F1, and AUC were selected as evaluation metrics. Moreover, we also chose alignment and uniformity to verify the effect of representation learning and UMAP (Uniform Manifold Approximation and Projection) to visualize fundus image embeddings. In summary, DR detection achieved state-of-the-art results across all metrics, with Accuracy = 98.91, Precision = 98.93, Recall = 98.90, F1 = 98.91, and AUC = 99.80. The grading also attained state-of-the-art results in terms of Accuracy and AUC, which were 85.61 and 93.97, respectively. The experimental results demonstrate that Angular Margin is an excellent intelligent medical diagnostic algorithm, performing well in both DR detection and grading tasks.

A crucial component of intuitive data visualization is presenting a hierarchical tree structure with interactive functions. For example, single-cell transcriptomics studies may generate gene expression values with developmental trajectories or cell lineage structures. Two common visualization methods, t-Distributed Stochastic Neighbor Embedding (t-SNE) and Uniform Manifold Approximation and Projection (UMAP), require two separate figures to depict the distribution of cell types and gene expression data, with low-dimension projections that may not capture the hierarchical structures among cells. Here, we present a JavaScript framework and an interactive web app named Collapsible Tree, which presents values jointly with interactive, expandable, and collapsible lineage structures. For example, the Collapsible Tree presents cellular states and gene expression from single-cell transcriptomics within a single hierarchical plot, enabling comparisons of gene expression across lineages and subtle patterns between sub-lineages. Our framework can facilitate the exploration of complicated value patterns that are not evident in UMAP or t-SNE plots. The Collapsible Tree web interface is available at https://collapsibletree.data2in.net. The JavaScript library source code is available at https://github.com/data2intelligence/collapsible_tree.

Comparison of dimension reduction techniques applied to the analysis of airborne radionuclide activity concentration