Róbert Pálovics
A5039678992- Data Stream Mining Techniques
- Single-cell and spatial transcriptomics
- Recommender Systems and Techniques
- Complex Network Analysis Techniques
- Advanced Bandit Algorithms Research
- Machine Learning and Data Classification
- Neuroinflammation and Neurodegeneration Mechanisms
- Peer-to-Peer Network Technologies
- Gene Regulatory Network Analysis
- Metabolomics and Mass Spectrometry Studies
- Statistical and Computational Modeling
- Human Mobility and Location-Based Analysis
- Advanced Graph Neural Networks
- Neural Networks and Applications
- Nutritional Studies and Diet
- Genetics, Aging, and Longevity in Model Organisms
- Adipose Tissue and Metabolism
- Epigenetics and DNA Methylation
- Opinion Dynamics and Social Influence
- Immune cells in cancer
- Erythrocyte Function and Pathophysiology
- Data Mining Algorithms and Applications
- Barrier Structure and Function Studies
- Cholesterol and Lipid Metabolism
- Peroxisome Proliferator-Activated Receptors
Stanford University
2018-2025
Neurosciences Institute
2020-2024
Laboratoire de Chimie
2023
Hungarian Academy of Sciences
2013-2017
Budapest University of Technology and Economics
2017
Institute for Computer Science and Control
2013-2016
Abstract Animal studies show aging varies between individuals as well organs within an individual 1–4 , but whether this is true in humans and its effect on age-related diseases unknown. We utilized levels of human blood plasma proteins originating from specific to measure organ-specific differences living individuals. Using machine learning models, we analysed 11 major estimated organ age reproducibly five independent cohorts encompassing 5,676 adults across the lifespan. discovered nearly...
Abstract Several genetic risk factors for Alzheimer’s disease implicate genes involved in lipid metabolism and many of these are highly expressed glial cells 1 . However, the relationship between glia pathology remains poorly understood. Through single-nucleus RNA sequencing brain tissue disease, we have identified a microglial state defined by expression droplet-associated enzyme ACSL1 with ACSL1-positive microglia being most abundant patients having APOE4/4 genotype. In human induced...
Abstract The human brain vasculature is of vast medical importance: its dysfunction causes disability and death, the specialized structure it forms—the blood-brain barrier—impedes treatment nearly all disorders. Yet, no molecular atlas exists. Here, we develop Vessel Isolation Nuclei Extraction for Sequencing (VINE-seq) to profile major vascular perivascular cell types through 143,793 single-nucleus transcriptomes from 25 hippocampus cortex samples 17 control Alzheimer’s disease (AD)...
Abstract Several genetic risk factors for Alzheimer’s Disease (AD) implicate genes involved in lipid metabolism and many of these are highly expressed glial cells. However, the relationship between glia AD pathology remains poorly understood. Through single-nucleus RNA-sequencing brain tissue, we have identified a microglial state defined by expression droplet (LD) associated enzyme ACSL1 with ACSL1-positive microglia most abundant patients APOE4/4 genotype. In human iPSC-derived (iMG)...
In this paper we give methods for time-aware music recommendation in a social media service with the potential of exploiting immediate temporal influences between users. We consider events when user listens to an artist first time and event follows some friend listening same short before. train blend matrix factorization that model relation influencer, influenced artist, both individual factor decompositions their weight learned by variants stochastic gradient descent (SGD). Special care is...
In this paper, we present algorithms that learn and update temporal node embeddings on the fly for tracking measuring similarity over time in graph streams. Recently, several representation learning methods have been proposed are capable of embedding nodes a vector space way captures network structure. Most known techniques extract from static snapshots. By contrast, modeling dynamics networks requires evolving representations. order to representations reflect changes local structure, rely...
Abstract Sociodemographic and lifestyle factors (sleep, physical activity, sedentary behavior) may predict obesity risk in early adolescence; a critical period during the life course. Analyzing data from 2971 participants (M = 11.94, SD 0.64 years) wearing Fitbit Charge HR 2 devices Adolescent Brain Cognitive Development (ABCD) Study, glass box machine learning models identified predictors Fitbit-derived measures of sleep, cardiovascular fitness, sociodemographic status. Key include...
The 2016 ACM Recommender Systems Challenge focused on the problem of job recommendations. Given a large dataset from XING that consisted anonymized user profiles, postings, and interactions between them, participating teams had to predict postings will interact with. challenge ran for four months with 366 registered teams. 119 those actively participated submitted together 4,232 solutions yielding in an impressive neck-and-neck race was decided within last days challenge.
In order to understand the cellular processes that underlie ageing, authors performed plasma proteomics at 10 different ages across lifespan of mouse. They integrated these data with a parallel large study published alongside this paper in same edition (1), which describes 'Mouse Ageing Cell Atlas', single-cell transcriptomic atlas characterizes changes gene expression age 23 tissues. Together, reveal clustered patterns ('trajectory groups') and protein levels, consistent coherent biological...
Abstract The aging process is universal, and it characterized by a progressive deterioration decrease in physiological function leading to decline on the organismal level. Nevertheless, number of genetic non-genetic interventions have been described, which successfully extend healthspan lifespan different species. Furthermore, clinical trials evaluating feasibility promote human health. goal annual Biological Sciences Section Gerontological Society America meeting was share current knowledge...
A plethora of centrality measures or rankings have been proposed to account for the importance nodes a network. In seminal study Boldi and Vigna (2014), comparative evaluation was termed difficult, arduous task. networks with fast dynamics, such as Twitter mention retweet graphs, predicting emerging is even more challenging. Our main result new, temporal walk based dynamic measure that models information propagation by considering order edge creation. Dynamic already started emerge in...
Aging is the single greatest cause of disease and death worldwide, so understanding associated processes could vastly improve quality life. While field has identified major categories aging damage such as altered intercellular communication, loss proteostasis, eroded mitochondrial function 1 , these deleterious interact with extraordinary complexity within between organs. Yet, a comprehensive analysis dynamics organism-wide lacking. Here we performed RNA-sequencing 17 organs plasma...
The area of online machine learning in big data streams covers algorithms that are (1) distributed and (2) work from with only a limited possibility to store past data. first requirement mostly concerns software architectures efficient algorithms. second one also imposes nontrivial theoretical restrictions on the modeling methods: In stream model, older is no longer available revise earlier suboptimal decisions as fresh arrives. this article, we provide an overview libraries well models for...
Activity-dependent neuroprotective protein (ADNP) mutations are linked with cognitive dysfunctions characterizing the autistic-like ADNP syndrome patients, who also suffer from delayed motor maturation. We thus hypothesized that is deregulated in versatile myopathies and local muscle deficiency results myopathy, treatable by fragment NAP. Here, single-cell transcriptomics identified as a major constituent of developing human muscle. transcript concentrations further predicted multiple...