Article
Computer Science, Interdisciplinary Applications
Nam D. Nguyen, Jiawei Huang, Daifeng Wang
Summary: The paper introduces an interpretable regularized learning model, deepManReg, for predicting phenotypes from multi-modal data. The model uses deep neural networks to learn cross-modal manifolds, aligns multi-modal features onto a common latent space, and prioritizes various multi-modal features and cross-modal interactions for phenotype prediction.
NATURE COMPUTATIONAL SCIENCE
(2022)
Article
Mathematics, Interdisciplinary Applications
Fuhong Min, Yizi Cheng, Lei Lu, Xinya Li
Summary: This paper introduces a novel memristive chaotic circuit based on two flux-controlled memristors, which is prone to extreme multistability. By establishing a third-order dimensionality reduction model, the internal dynamics of the circuit is investigated and validated through simulation, showing its potential for both research and practical applications.
INTERNATIONAL JOURNAL OF BIFURCATION AND CHAOS
(2021)
Article
Biochemical Research Methods
Omid Bazgir, Souparno Ghosh, Ranadip Pal
Summary: The study focuses on predicting anti-cancer drug sensitivity in individual cell lines using deep learning models, specifically the REFINED-CNN model. By considering an ensemble of REFINED-CNN models built under different distance metrics and projection schemes, the study demonstrates significant improvements in prediction performance compared to individual models. The research also develops a theoretical framework for combining different distance metrics to achieve a single 2D mapping, showing that distance-averaged REFINED-CNN produces comparable performance with lower computational cost than stacking REFINED-CNN ensemble.
Article
Biology
Yongjie Xu, Zelin Zang, Jun Xia, Cheng Tan, Yulan Geng, Stan Z. Li
Summary: This paper proposes a general visualization method called deep visualization (DV) that can preserve the inherent structure of data and handle batch effects in various datasets. DV learns a structure graph to describe the relationships between data samples and transforms the data into a visualization space while preserving the geometric structure and correcting batch effects.
COMMUNICATIONS BIOLOGY
(2023)
Article
Environmental Sciences
Matthew C. Cave, Christina M. Pinkston, Shesh N. Rai, Banrida Wahlang, Marian Pavuk, Kimberly Z. Head, Gleta K. Carswell, Gail M. Nelson, Carolyn M. Klinge, Douglas A. Bell, Linda S. Birnbaum, Brian N. Chorley
Summary: This study suggests that environmental exposure to PCBs is associated with liver injury in humans and provides insights into the potential modes of PCB action. MiR-derived liquid liver biopsy represents a promising new technique for environmental hepatology cohort studies.
ENVIRONMENTAL HEALTH PERSPECTIVES
(2022)
Review
Genetics & Heredity
Ziye Zhao, Wen Yang, Yixiao Zhai, Yingjian Liang, Yuming Zhao
Summary: The exploration of DNA-binding proteins is crucial in studying biological life activities, and machine learning algorithms have shown excellent performance in detecting DBPs. Our method, using feature extraction and the XGBoost model, achieves better results with high accuracy and simplicity compared to other methods.
FRONTIERS IN GENETICS
(2022)
Article
Biochemical Research Methods
Shuyi Zhang, Jacob R. Leistico, Raymond J. Cho, Jeffrey B. Cheng, Jun S. Song
Summary: This study introduces two spectral algorithms on multilayer graphs for clustering cells in multi-omic single-cell sequencing datasets, demonstrating the WLL method as a new spectral graph theoretic reformulation of the popular Seurat weighted nearest neighbor algorithm.
Article
Mathematics, Interdisciplinary Applications
Mujtaba Husnain, Malik Muhammad Saad Missen, Shahzad Mumtaz, Dost Muhammad Khan, Mickael Coustaty, Muhammad Muzzamil Luqman, Jean-Marc Ogier, Hizbullah Khattak, Sikandar Ali, Ali Samad
Summary: This research utilizes t-SNE to process high-dimensional data of Urdu handwritten characters and numerals, achieving high recognition accuracy. The novel approach of using reduced dimensions for character-level recognition reduces computation time.
Article
Biochemical Research Methods
Kai Cao, Yiguang Hong, Lin Wan
Summary: Motivated by the need for effective approaches to integrate single-cell multi-omics data, this study presents Pamona, a partial Gromov-Wasserstein distance-based manifold alignment framework. It aims to delineate and represent the shared and dataset-specific cellular structures across modalities. Pamona demonstrates superior performance in accurately identifying shared and dataset-specific cells, recovering and aligning cellular structures, outperforming existing methods. The framework also allows for the incorporation of prior information to enhance alignment quality.
Article
Computer Science, Software Engineering
Tacito Trindade de Araujo Tiburtino Neves, Rafael Messias Martins, Danilo Barbosa Coimbra, Kostiantyn Kucher, Andreas Kerren, Fernando Paulovich
Summary: Streaming data applications are becoming more common, but existing visualization methods have limitations. This paper presents a novel incremental dimensionality reduction technique called Xtreaming, which can continuously update visual representations without revisiting the data.
COMPUTERS & GRAPHICS-UK
(2022)
Article
Microbiology
George Armstrong, Cameron Martino, Gibraan Rahman, Antonio Gonzalez, Yoshiki Vazquez-Baeza, Gal Mishne, Rob Knight
Summary: This study discusses the advantages and limitations of using UMAP for dimensionality reduction in microbiome data, demonstrating that UMAP can improve cluster representation, enhance correlation with biological variation gradients, and provide parameter recommendations for preserving global geometry. UMAP is recommended as a complementary visualization method for microbiome beta diversity studies.
Article
Biochemical Research Methods
Tobias Wangberg, Joanna Tyrcha, Chun-Biu Li
Summary: The proposed shape-aware stochastic neighbor embedding method outperforms t-SNE, UMAP, and PHATE in visualizing imbalanced, nonlinear, continuous, and hierarchically structured data. Additionally, the method can automatically choose the optimal hyper-parameter in a data-driven way, consistently performing well across different test cases.
BMC BIOINFORMATICS
(2022)
Article
Mathematics, Applied
K. Gajamannage, D. I. Jayathilake, Y. Park, E. M. Bollt
Summary: This paper discusses the limitations of traditional statistical methods for solving spatiotemporal dynamical systems and proposes artificial neural networks (ANNs) as a better approach. Specifically, recurrent neural networks (RNNs) are introduced as a type of ANN capable of processing variable-length input sequences, making them applicable for a wide range of problems in spatiotemporal dynamical systems. The paper analyzes the performance of RNNs in three tasks, including repairing erroneous Lorenz equations, repairing damaged collective motion trajectories, and forecasting streamflow time series with spikes, demonstrating the broad applicability of RNNs in reconstructing and forecasting the dynamics of dynamical systems.
Article
Chemistry, Physical
Zixin Zhuang, Amanda S. Barnard
Summary: This article introduces a structure-free encoding method called Mendeleev encoding for materials. The evaluation of Mendeleev encoding on three data sets for battery applications shows that it is more accurate, stable, and reliable than alternative structure-free encoding methods, and consistently provides superior clustering results.
CHEMISTRY OF MATERIALS
(2023)
Article
Computer Science, Information Systems
Naoual Nassiri, Abdelhak Lakhouaja, Violetta Cavalli-Sforza
Summary: Readability is a measure of how well a written text matches a reader's skill or grade level. Evaluating the readability of second or foreign language (L2) texts is crucial. Arabic, in particular, poses challenges due to its complexity, highlighting the need for an accurate readability measure. In this paper, the authors present an approach to automatically measure the readability of Arabic as a foreign language. Through experiments, they identified the most relevant features and achieved an L2 readability accuracy of 86.15%.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Neurosciences
David Pamies, Tatjana Vujic, Domitille Schvartz, Julien Boccard, Cendrine Repond, Carolina Nunes, Serge Rudaz, Jean-Charles Sanchez, Victor Gonzalez-Ruiz, Marie-Gabrielle Zurich
Summary: This study found that the activation of human ReN-derived astrocytes is closely related to energy metabolism, and increased glycolysis may be considered as an endpoint for detecting astrocyte activation by potentially neurotoxic compounds in vitro. This is important for studying the mechanisms of neurotoxicity and the direct effects of chemicals on astrocytes.
MOLECULAR NEUROBIOLOGY
(2023)
Correction
Medicine, General & Internal
Jean Terrier, Frederic Gaspar, Pierre Fontana, Youssef Daali, Jean-Luc Reny, Chantal Csajka, Caroline F. Samer
AMERICAN JOURNAL OF MEDICINE
(2023)
Article
Chemistry, Analytical
Gioele Visconti, Julien Boccard, Max Feinberg, Serge Rudaz
Summary: In the past two decades, liquid chromatography coupled to mass-spectrometry (LC-MS) has become the gold standard for qualitative and quantitative analysis of small molecules. The design of calibration curves has evolved with instrumental advances, introducing innovative approaches to improve accuracy and efficiency. This tutorial covers the advances in LC-MS quantitative analysis for small molecules in complex matrices, including fundamental aspects in calibration, modern methodologies, and applications. The importance of international guidelines for analytical method validation that align with calibration strategies and analytical instrumentation is emphasized.
ANALYTICA CHIMICA ACTA
(2023)
Article
Pharmacology & Pharmacy
Christian Skalafouris, Anne-Laure Blanc, Olivier Grosgurin, Christophe Marti, Caroline Samer, Christian Lovis, Pascal Bonnabry, Bertrand Guignard
Summary: The study developed electronic queries to assist pharmacists in medication reviews and assessed their performance in detecting drug-related problems (DRPs). The results showed that these electronic queries had high sensitivity and negative predictive value, which can contribute to improving clinical decision support systems.
INTERNATIONAL JOURNAL OF CLINICAL PHARMACY
(2023)
Article
Ethics
Cristina Bosmani, Sonia Carboni, Caroline Samer, Christian Lovis, Thomas Perneger, Angela Huttner, Bernard Hirschel
Summary: In this study, consent bias was assessed in a cohort of over 40,000 adult patients asked to provide consent for the use of their clinical and biological data. The results showed that older patients with more comorbidities and Swiss nationality were more likely to provide consent, indicating a potential bias in actively seeking consent and compromising the external validity of data obtained.
BMC MEDICAL ETHICS
(2023)
Letter
Toxicology
Daniela Pelclova, Karel Lach
CLINICAL TOXICOLOGY
(2023)
Article
Biochemical Research Methods
Sergey Girel, Davy Guillarme, Szabolcs Fekete, Serge Rudaz, Victor Gonzalez-Ruiz
Summary: Monitoring the central carbon metabolism network using LC-MS analysis is challenging due to the diverse chemical nature of analytes and non-specific adsorption on metal surfaces. In this study, different chromatographic methods were investigated for untargeted analysis of energy metabolism-related analytes in biological samples, and the best performance was achieved with sulfoalkylbetaine HILIC. Additionally, an extra hydrophilic modulation may be necessary for better resolution of carboxylic acids in zHILIC mode.
JOURNAL OF CHROMATOGRAPHY A
(2023)
Article
Pharmacology & Pharmacy
Frederique Rodieux, Youssef Daali, Victoria Rollason, Caroline F. Samer, Kuntheavy Ing Lorenzini
Summary: Pharmacokinetics in children is highly variable due to various factors including ontogeny, pharmacogenetics, gender, comorbidities, and drug-drug interactions. Understanding the impact of genetic polymorphisms on drug metabolism and response is lacking in pediatric population. This study aimed to investigate the use of cytochromes P450 (CYP) genotyping and phenotyping tests in children and assess the correlation between genotype and phenotype.
FRONTIERS IN PHARMACOLOGY
(2023)
Article
Pharmacology & Pharmacy
Ali El Rida El Masri, Caroline Tobler, Breunis Willemijn, Andre O. Von Bueren, Marc Ansari, Caroline Flora Samer
Summary: Methotrexate is an immunosuppressant and chemotherapeutic agent used for treating autoimmune disorders and cancers. Its main adverse effects are bone marrow suppression and gastrointestinal complications, but hepatotoxicity and nephrotoxicity are also common.
FRONTIERS IN PHARMACOLOGY
(2023)
Review
Genetics & Heredity
Aurelien Simona, Wenyu Song, David W. Bates, Caroline Flora Samer
Summary: Pharmacogenomics (PGx) aims to personalize drug therapy based on patient's genetic makeup. Polygenic risk scores (PRS) have emerged as a promising tool to consider the complex interplay and polygenic nature of patients' genetic predisposition affecting drug response. However, there is still a need to demonstrate the clinical utility and implementation of PRS in daily care, and collaboration between bioinformatician, treating physicians, and genetic consultants is crucial for transparent and trustworthy integration of PRS results into real-world medical decisions.
FRONTIERS IN GENETICS
(2023)
Article
Chemistry, Analytical
Miguel de Figueiredo, Serge Rudaz, Julien Boccard
Summary: This study proposes a new method to handle unbalanced groups in multifactorial experimental designs. The method extends a prior rebalancing strategy and uses ANOVA for decomposition, while preserving within-group variation and orthogonal effect matrices, avoiding confusion of effects and improving model interpretation. It can be combined with any multivariate method for high-dimensional data analysis.
ANALYTICA CHIMICA ACTA
(2023)
Article
Chemistry, Analytical
Gioele Visconti, Miguel de Figueiredo, Oriane Strassel, Julien Boccard, Nicolas Vuilleumier, David Jaques, Belen Ponte, Serge Rudaz
Summary: The study proposes a multitargeted internal calibration (MTIC) approach for accurate quantitative analysis in liquid chromatography-mass spectrometry (LC-MS) for endogenous compound quantification when there is no blank matrix available. The approach integrates stable isotope labeled (SIL) standards as internal calibrants, allowing simultaneous quantification of analytes using within sample calibration. The developed MTIC workflow demonstrates reliable quantification of metabolites related to chronic kidney disease (CKD) in small serum or plasma samples.
ANALYTICAL CHEMISTRY
(2023)
Article
Chemistry, Analytical
William Bello, Julian Pezzatti, Markoulina Berger-Gryllaki, Serge Rudaz, Farshid Sadeghipour
Summary: Prefilled plastic packaging in hospital pharmacy is beneficial for preventing waste, errors, contamination, and accidents. A new sensitive method based on LC-HRMS has been developed to analyze plastic additives in the prefilled drug solutions prepared in hospital pharmacies. Plastic additives can be rapidly identified and their toxicology can be assessed using this method.
JOURNAL OF PHARMACEUTICAL AND BIOMEDICAL ANALYSIS
(2023)