A5088622606- Meteorological Phenomena and Simulations
- Climate variability and models
- Oceanographic and Atmospheric Processes
- Genomics and Chromatin Dynamics
- Atmospheric and Environmental Gas Dynamics
- RNA modifications and cancer
- Hydrological Forecasting Using AI
- Energy Load and Power Forecasting
- Epigenetics and DNA Methylation
- Stock Market Forecasting Methods
- Image and Signal Denoising Methods
- Reservoir Engineering and Simulation Methods
- Cryospheric studies and observations
- Tropical and Extratropical Cyclones Research
- Computational Physics and Python Applications
- Arctic and Antarctic ice dynamics
- Nuclear Engineering Thermal-Hydraulics
- Digital and Cyber Forensics
- Simulation and Modeling Applications
- Coastal and Marine Dynamics
- Ocean Waves and Remote Sensing
- Chromosomal and Genetic Variations
- Advanced Image Processing Techniques
- Innovative Educational Techniques
- Underwater Vehicles and Communication Systems
Fudan University
2023-2024
Tongji University
2012-2023
Polar Research Institute of China
2023
Dana-Farber Cancer Institute
2012-2013
Harvard University
2013
University of Southern California
2012
Tân Tạo University
2006
A large collection of new modENCODE and ENCODE genome-wide chromatin data sets from cell lines developmental stages in worm, fly human are analysed; this reveals many conserved features organization among the three organisms, as well notable differences composition locations repressive chromatin. This study describes numerous Homo sapiens, Drosophila melanogaster Caenorhabditis elegans generated by consortia. The results point to while identifying Genome function is dynamically regulated...
Abstract. The El Niño–Southern Oscillation (ENSO) is an extremely complicated ocean–atmosphere coupling event, the development and decay of which are usually modulated by energy interactions between multiple physical variables. In this paper, we design a multivariate air–sea coupler (ASC) based on graph using features On basis coupler, ENSO deep learning forecast model (named ENSO-ASC) proposed, whose structure adapted to characteristics dynamics, including encoder decoder for capturing...
El Niño-Southern Oscillation (ENSO) is the dominant atmosphere-ocean coupled mode of year-to-year variations in tropical Pacific. It shows diverse spatiotemporal characteristics and casts major influences on seasonal predictions global weather-climate extrema. Despite numerous dynamical statistical models for ENSO prediction predictability studies, they are commonly subjected to one-to-three issues among less skillful simulation Niño diversity, huge requirements...
Abstract El Niño‐Southern Oscillation (ENSO) is the dominant atmosphere–ocean coupled mode of year‐to‐year variations in tropical Pacific. It shows diverse spatiotemporal characteristics and casts major influences on seasonal predictions global weather–climate extrema. Despite numerous dynamical statistical models for ENSO prediction predictability studies, they are commonly subjected to one‐to‐three issues among less skillful simulation Niño diversity, huge requirements computational...
Abstract Summary: Transcription and chromatin regulators, histone modifications play essential roles in gene expression regulation. We have created CistromeMap as a web server to provide comprehensive knowledgebase of all the publicly available ChIP-Seq DNase-Seq data mouse human. also manually curated metadata ensure annotation consistency, developed user-friendly display matrix for quick navigation retrieval specific factors, cells papers. Finally, we users with summary statistics studies....
Abstract Continuing the tropical Pacific multivariate air‐sea coupler proposed by us before, we design Global spatial‐temporal Teleconnection Coupler (GTC), which is modeled to discover latent teleconnections among global sea surface temperature (SST). To this end, Pacific, Indian, and Atlantic oceans are divided into small ocean patches that compose a dynamics graph, in adjacent relationships artificially constructed prior knowledge non‐adjacent learned from data deep learning methods....
In this paper, a Physics-Informed Red Tide forecast model (PIRT) considering causal-inferred predictors selection is proposed. Specifically, the directed acyclic graph-graph neural network (DAG-GNN) method first applied to quantify causality among multiple ocean-atmosphere-biology variables for selecting most significant of red tides (or other chlorophyll variations). Then, encoder-decoder consisting an Energy Attention Module (EAM) built daily tide forecasting. The multi-sourced...
Advances in deep learning have created new opportunities for improving traditional numerical models. As the radiation parameterization scheme is crucial and time-consuming models, researchers sought to replace it with emulators. However, progress has been hindered at offline emulation stage due technical complexity of implementation. Additionally, performance emulators when coupled large-scale models yet be verified. In this paper, we developed a tool called Fortran Torch Adaptor (FTA)...
Abstract. The global impact of an El Niño–Southern Oscillation (ENSO) event can differ greatly depending on whether it is eastern Pacific (EP)-type or a central (CP)-type event. Reliable predictions the two types ENSO are therefore critical importance. Here we construct deep neural network with multichannel structure for (named ENSO-MC) to simulate spatial evolution sea surface temperature (SST) anomalies events. We select SST, heat content and wind stress (i.e., three key ingredients...
Abstract Summary: Chromatin immunoprecipitation and DNase I hypersensitivity assays with high-throughput sequencing have greatly accelerated the understanding of transcriptional epigenetic regulation, although data reuse for community experimental biologists has been challenging. We created a portal CistromeFinder that can help query, evaluate visualize publicly available in human mouse. The database currently contains 6378 samples over 4391 datasets, 313 factors 102 cell lines or...
The North Atlantic Oscillation (NAO) is a major climatic phenomenon in the Northern Hemisphere, but underlying air–sea interaction and physical mechanisms remain elusive. Despite successful short-term forecasts using physics-based numerical models, longer-term of NAO continue to pose challenge. In this study, we employ advanced data-driven causal discovery techniques explore causality between multiple ocean–atmosphere processes NAO. We identify best predictors based on analysis develop...
Climate downscaling is a way to provide finer resolution data at local scales, which has been widely used in meteorological research. The two main approaches for climate are dynamical and statistical. traditional methods quite time- resource-consuming based on general circulation models (GCMs). Recently, more researchers construct statistical deep learning model motivated by the single-image superresolution (SISR) process computer vision (CV). This an approach that uses historical...
Existing convolution recurrent neural networks (ConvRNNs)-based memory cells majorly take advantage of gated structures and attention mechanisms to extract discontinuous latent associations for spatial-temporal sequence forecast (STSF) problems, which may lead serious over-fitting spurious relationships with correlated noise. It is a consensus that incorporating cause-effect in modeling can alleviate these problems. In this paper, we propose Causality Attention Unit (CAU) assist ConvRNNs by...
Satellite-based remote sensing technology plays a significant role in identifying tropical cyclones (TCs), and most of the current research focuses on intensity estimation. However, analyzing wind structure TCs, which is directly related to danger they bring, remains challenge. By adding prior knowledge TCs into model, we propose physics-incorporated network based multi-task learning estimate radii intensity, whose layers can automatically extract rotation-invariant features TC core from...
This paper proposes a global sea surface temperature (SST) bias correction and downscaling unifying model based on the generative adversarial network. The generator of uses module to correct numerical forecasting results. Then it reusable shared improve resolution corrected data gradually. discriminator evaluates quality results as criterion for training. And physics-informed dynamics penalty term is included in loss function performance model. Based 1°-resolution SST GFDL SPEAR (Seamless...
The downscaling of climate data has always been a hotspot in meteorological research. traditional method is to adjust numerical models simulate high-resolution situation, which quite time and resource consuming. Recently, deep learning methods have provided an appreciable new insight successfully downscale multiple phenomena. However, existing simply treat problem as image super-resolution extract the spatial features from high resolution (HR) directly, ignores detailed physical varieties...
El Niño-Southern Oscillation (ENSO) events have significant impacts on global climate change, and the research their accurate forecasting dynamic predictability holds remarkable scientific engineering values. Recent years, we constructed two ENSO deep learning models, ENSO-ASC ENSO-GTC, which are both incorporated with prior mechanisms. Specifically, former possesses multivariate air-sea coupler (ASC), can simulate occurrence decay of events, accompanied by concurrent energy...