Results 171 to 180 of about 283,214 (276)

Hierarchical Prediction and Perturbation of Chromatin Organization Reveal How Loop Domains Mediate Higher‐Order Architectures

open access: yesAdvanced Science, EarlyView.
HiCGen introduces a hierarchical deep learning framework to predict genome organization across spatial scales using DNA sequences and genomic features. The model enables cross‐cell‐type predictions and in silico perturbation analysis, revealing correlations between loop domains and higher‐order structures.
Jiachen Wei, Yue Xue, Yi Qin Gao
wiley   +1 more source

Topological momentum skyrmions in Mie scattering fields. [PDF]

open access: yesNanophotonics
Chen P, Lee KX, Meiler TC, Shen Y.
europepmc   +1 more source

Q‐GEM: Quantum Chemistry Knowledge Fusion Geometry‐Enhanced Molecular Representation for Property Prediction

open access: yesAdvanced Science, EarlyView.
This work introduces E‐GeoGNN, a graph neural network framework that systematically encodes molecular structural information through three hierarchical correlation graphs: atom‐bond graph, bond‐angle graph, and angle‐dihedral graph. By incorporating quantum and geometric dual‐scale self‐supervised pretraining, Q‐GEM is proposed, a novel molecular ...
Zhijiang Yang   +7 more
wiley   +1 more source

Unusual topological polar texture in moiré ferroelectrics. [PDF]

open access: yesNat Commun
Li Y   +10 more
europepmc   +1 more source

Spin‐Orbit Angular Momentum Conversion in Coupled Nonparaxial Bessel Acoustic Vortices

open access: yesAdvanced Science, EarlyView.
This study gives a formulaic description of the spin‐orbit angular momentum (AM) conversion in coupled fields of two nonparaxial Bessel acoustic vortices. Experimental results of wave field measurements provide compelling evidence for spin‐orbit AM conversion via passive artificial structures.
Di‐Chao Chen   +5 more
wiley   +1 more source

SGCD: High‐Resolution Spatial Domain Characterization via Data Interpolation and Cell‐Type Deconvolution

open access: yesAdvanced Science, EarlyView.
SGCD presents a novel approach for tissue spatial domain identification by employing interpolation to estimate inter‐spot gene expression and deconvolution to resolve cell‐type composition in both sampled and interstitial regions. By integrating gene expression, cell type, and spatial coordinates within a graph contrastive learning framework, SGCD ...
Tianjiao Zhang   +7 more
wiley   +1 more source

MGPT: A Multi‐task Graph Prompt Learning Framework for Drug Discovery

open access: yesAdvanced Science, EarlyView.
MGPT is a unified multi‐task graph prompt learning model providing generalizable and robust graph representations for few‐shot drug association prediction. MGPT demonstrates the ability of seamless task switching and outperforms competitive approaches in few‐shot scenarios.
Yang Li   +4 more
wiley   +1 more source

Home - About - Disclaimer - Privacy