Caption Royale: Exploring the Design Space of Affective Captions from the Perspective of Deaf and Hard-of-Hearing Individuals

Опубликована: Май 11, 2024

Affective captions employ visual typographic modulations to convey a speaker's emotions, improving speech accessibility for Deaf and Hard-of-Hearing (dhh) individuals. However, the most effective expressing emotions remain uncertain. Bridging this gap, we ran three studies with 39 dhh participants, exploring design space of affective captions, which include parameters like text color, boldness, size, so on. Study 1 assessed preferences nine these styles, each conveying either valence or arousal separately. 2 combined 1's top-performing styles measured depicting both simultaneously. Participants outlined readability, minimal distraction, intuitiveness, emotional clarity as key factors behind their choices. In 3, an emotion-recognition task were used compare how 2's winning performed versus non-styled baseline. Based on our findings, present two best-performing recommendations applications employing captions.

Язык: Английский

---

Metabolic Connectome and Its Role in the Prediction, Diagnosis, and Treatment of Complex Diseases

Metabolites, Год журнала: 2024, Номер 14(2), С. 93 - 93

Опубликована: Янв. 26, 2024

The interconnectivity of advanced biological systems is essential for their proper functioning. In modern connectomics, entities such as proteins, genes, RNA, DNA, and metabolites are often represented nodes, while the physical, biochemical, or functional interactions between them edges. Among these entities, particularly significant they exhibit a closer relationship to an organism’s phenotype compared genes proteins. Moreover, metabolome has ability amplify small proteomic transcriptomic changes, even those from minor genomic changes. Metabolic networks, which consist complex comprising hundreds interactions, play critical role in research by mediating energy conversion chemical reactions within cells. This review provides introduction common metabolic network models construction methods. It also explores diverse applications networks elucidating disease mechanisms, predicting diagnosing diseases, facilitating drug development. Additionally, it discusses potential future directions networks. Ultimately, this serves valuable reference researchers interested modeling, analysis, applications.

Язык: Английский

---

Improved distance correlation estimation

Applied Intelligence, Год журнала: 2025, Номер 55(4)

Опубликована: Янв. 3, 2025

Distance correlation is a novel class of multivariate dependence measure, taking positive values between 0 and 1, applicable to random vectors not necessarily equal arbitrary dimensions. It offers several advantages over the well-known Pearson coefficient, most important being that distance equals zero if-and-only if- are independent. There two different estimators available in literature. The first estimator, proposed by Székely et al. (Ann Stat 35:2769–279 2007), based on an asymptotically unbiased estimator covariance, which V-statistic. second builds covariance Rizzo (Stat 42:2382–2412 2014), shown be U-statistic Huo (Technometrics 58:435–447 2016). This study evaluates their efficiency (mean squared error) compares computational times for both methods under structures. Under conditions independence or near-independence, V-estimates biased, while U-estimator frequently cannot computed due negative values. To address this challenge, convex linear combination former studied, yielding good results regardless level dependence. Additionally, medical database studied discussed.

Язык: Английский

---

Transcriptomic signatures and network‐based methods uncover new senescent cell anti‐apoptotic pathways and senolytics

FEBS Journal, Год журнала: 2025, Номер unknown

Опубликована: Янв. 27, 2025

Cellular senescence is an irreversible cell cycle arrest caused by various stressors that damage cells. Over time, senescent cells accumulate and contribute to the progression of multiple age‐related degenerative diseases. It believed these partly due their ability evade programmed death through development activation survival antiapoptotic resistance mechanisms; however, many aspects how mechanisms develop activate are still unknown. By analyzing transcriptomic signature profiles generated LINCS L1000 project using network‐based methods, we identified genes could represent new senescence‐related mechanisms. Additionally, employing same methodology, over 600 molecules with potential senolytic activity. Experimental validation our computational findings confirmed activity Fluorouracil, whose would be mediated a multitarget mechanism, revealing its targets AURKA, EGFR, IRS1, SMAD4, KRAS pathways (SCAPs). The depend on stimulus induces cellular senescence. SCAP proposed in this work offer insights into survive. Identifying drugs paves way for developing pharmacological therapies eliminate selectively.

Язык: Английский

---

Machine Learning-assisted Prediction of Organic Solar Cell Efficiency from TCA triplelayer reflectance Spectra

Optics Communications, Год журнала: 2025, Номер unknown, С. 131654 - 131654

Опубликована: Фев. 1, 2025

Язык: Английский

---

Modeling Potential Habitats of Macrophytes in Small Lakes: A GIS and Remote Sensing-Based Approach

Remote Sensing, Год журнала: 2024, Номер 16(13), С. 2339 - 2339

Опубликована: Июнь 26, 2024

Macrophytes, which are foundational to freshwater ecosystems, face significant threats due habitat degradation globally. Habitat suitability models vital tools used investigate the relationship between macrophytes and their environment. This study addresses a critical gap by developing Geographic information system-based HSM tailored for small lakes, often overlooked in ecological studies. We included various abiotic predictors model potential macrophyte several lakes southern Bavaria (Germany). Key factors such as distance groundwater inflow, depth, availability of photosynthetically active radiation (PAR), littoral slope were identified occurrence. Notably, integrates remote sensing-based data derive PAR at growing depths using Sentinel-2 MSI data. Integration an MSI-based time series enabled introduction temporal component allowing monitoring predicting changes habitats over time. The modeled score correlates highly (R = 0.908) with see great promise modeling tool water management; particular, use holds advancing management. By demonstrating efficacy GIS- HSM, we pave way future applications this innovative approach conservation resource

Язык: Английский

---

Enhancing the Vietoris–Rips simplicial complex for topological data analysis: applications in cancer gene expression datasets

International Journal of Data Science and Analytics, Год журнала: 2024, Номер unknown

Опубликована: Апрель 16, 2024

Abstract The aim of this study is to enhance the extraction informative features from complex data through application topological analysis (TDA) using novel overlapping measures. Topological has emerged as a promising methodology for extracting meaningful insights datasets. Existing approaches in TDA often involve extrapolating points distance correlation measures, which subsequently constrain downstream predictive tasks. Our objective improve construction Vietoris–Rips simplicial by introducing These measures take into account interplay direct connection strengths and shared neighbours, leading identification persistent features. We propose utilisation optimise complex, offering more refined representation structures. results plentiful This enhancement contributes an improvement up 20% cancer phenotype prediction across different types. demonstrates effectiveness utilising optimising complex. identified significantly accuracy phenotypes. approach potential advance field our understanding structures, particularly context research modelling. Further exploration these may yield valuable various domains dealing with intricate

Язык: Английский

---

Transcriptomic signatures and network-based methods uncover new Senescent Cell Anti-Apoptotic Pathways and Senolytics

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2024, Номер unknown

Опубликована: Май 30, 2024

Abstract Cellular senescence is an irreversible cell cycle arrest caused by various stressors that damage cells. Over time, senescent cells accumulate and contribute to the progression of multiple age-related degenerative diseases. It believed these partly due their ability evade programmed death through development activation survival anti-apoptotic resistance mechanisms; however, many aspects how mechanisms develop activate are still unknown. By analyzing transcriptomic signature profiles generated LINCS L1000 project using network-based methods, we identified genes could represent new senescence-related mechanisms. Additionally, employing same methodology, over 600 molecules with potential senolytic activity. Experimental validation our computational findings confirmed activity Fluorouracil, whose would be mediated a multi-target mechanism, revealing its targets AURKA, EGFR, IRS1, SMAD4, KRAS senescence-associated pathways. The pathways depend on stimulus induces cellular senescence. SCAPs proposed in this work offer insights into survive. Identifying drugs paves way for developing pharmacological therapies eliminate selectively. Graphical

Язык: Английский

---

SAE-Impute: imputation for single-cell data via subspace regression and auto-encoders

BMC Bioinformatics, Год журнала: 2024, Номер 25(1)

Опубликована: Окт. 1, 2024

Язык: Английский

---

scMultiNODE: Integrative Model for Multi-Modal Temporal Single-Cell Data

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2024, Номер unknown

Опубликована: Окт. 29, 2024

Abstract Measuring single-cell genomic profiles at different timepoints enables our understanding of cell development. This is more comprehensive when we perform an integrative analysis multiple measurements (or modalities) across various developmental stages. However, obtaining such from the same set single cells resource-intensive, restricting ability to study them integratively. We propose unsupervised integration model, scMultiNODE, that integrates gene expression and chromatin accessibility in developing while preserving type variations cellular dynamics. scMultiNODE uses autoencoders learn nonlinear low-dimensional representation optimal transport align measurements. Next, it utilizes neural ordinary differential equations explicitly model development with a regularization term dynamic latent space. Our experiments on four real-world datasets show can integrate temporally profiled multi-modal better than existing methods focus tend ignore also scMultiNODE’s joint space helps downstream Availability The data code are publicly available https://github.com/rsinghlab/scMultiNODE .

Язык: Английский

---