Article
|
Open Access
Featured
-
-
Article
| Open AccessCAJAL enables analysis and integration of single-cell morphological data using metric geometry
Cell morphology is one of the most described phenotypes in biology, yet systematic quantification and classification of morphology remains limited. Here, the authors present a computational approach for cell morphometry and multi-modal analysis based on concepts from metric geometry.
- Kiya W. Govek
- , Patrick Nicodemus
- & Pablo G. Camara
-
Article
| Open AccessNon-line-of-sight imaging with arbitrary illumination and detection pattern
The authors propose a confocal complemented signal-object collaborative regularization method for non-line-of-sight (NLOS) imaging without specific requirements on the spatial pattern of measurement points. The method extends the application range of NLOS imaging.
- Xintong Liu
- , Jianyu Wang
- & Lingyun Qiu
-
Article
| Open AccessRetention time prediction for chromatographic enantioseparation by quantile geometry-enhanced graph neural network
Chromatographic enantioseparation requires tedious trials to find proper experimental conditions. Here, the authors construct a deep learning model to predict retention times of chiral molecules and obtain the separation probability under given conditions.
- Hao Xu
- , Jinglong Lin
- & Fanyang Mo
-
Article
| Open AccessData-driven prediction of complex crystal structures of dense lithium
Recent experiments reveal undetermined crystalline phases near the melting minimum region in lithium. Here, the authors use a crystal structure search method combined with machine learning to explore the energy landscape of lithium and predict complex crystal structures.
- Xiaoyang Wang
- , Zhenyu Wang
- & Yanming Ma
-
Article
| Open AccessQuantum process tomography with unsupervised learning and tensor networks
In quantum technologies, scalable ways to characterise errors in quantum hardware are highly needed. Here, the authors propose an approximate version of quantum process tomography based on tensor network representations of the processes and data-driven optimisation.
- Giacomo Torlai
- , Christopher J. Wood
- & Leandro Aolita
-
Article
| Open AccessEmergent geometry and duality in the carbon nucleus
Carbon (12C) nucleus has interesting characteristics including the existence of the Hoyle state. Here the authors discuss the structure of the nuclear states of 12C by using nuclear lattice effective field theory.
- Shihang Shen
- , Serdar Elhatisari
- & Ulf-G. Meißner
-
Article
| Open AccessExplainable multi-task learning for multi-modality biological data analysis
Multimodal biological data is challenging to analyze. Here, the authors develop UnitedNet, an explainable deep neural network for analyzing single-cell multimodal biological data and estimating relationships between gene expression and other modalities with cell-type specificity.
- Xin Tang
- , Jiawei Zhang
- & Jia Liu
-
Article
| Open AccessMachine learning prediction of the degree of food processing
Evidence suggests that increased consumption of ultra-processed food has adverse health implications, however, it remains difficult to classify processed food. Here, the authors introduce FPro, a machine learning-based score predicting the degree of food processing.
- Giulia Menichetti
- , Babak Ravandi
- & Albert-László Barabási
-
Article
| Open AccessAutomated optimisation of solubility and conformational stability of antibodies and proteins
Antibodies find key applications in research, diagnostics, and therapeutics, but their development can be impeded by poor stability or solubility. Here the authors developed a computational strategy that enables antibody optimisation, without affecting functionality.
- Angelo Rosace
- , Anja Bennett
- & Pietro Sormanni
-
Article
| Open AccessAtomic-scale origin of the low grain-boundary resistance in perovskite solid electrolyte Li0.375Sr0.4375Ta0.75Zr0.25O3
Oxide solid electrolytes generally suffer from high grain boundary resistance. Here, the authors use advanced electron microscopy, along with an active learning moment tensor potential, to reveal the atomic-scale origin of low grain-boundary resistance in Li0.375Sr0.4375Ta0.75Zr0.25O3.
- Tom Lee
- , Ji Qi
- & Xiaoqing Pan
-
Article
| Open AccessCellcano: supervised cell type identification for single cell ATAC-seq data
Accurately annotating cell types is a fundamental step in single-cell omics data analysis. Here, the authors develop a computational method called Cellcano based on a two-round supervised learning algorithm to identify cell types for scATAC-seq data and perform benchmarking to demonstrate its accuracy, robustness and computational efficiency.
- Wenjing Ma
- , Jiaying Lu
- & Hao Wu
-
Article
| Open AccessMeta-learning biologically plausible plasticity rules with random feedback pathways
The biological plausibility of backpropagation and its relationship with synaptic plasticity remain open questions. The authors propose a meta-learning approach to discover interpretable plasticity rules to train neural networks under biological constraints. The meta-learned rules boost the learning efficiency via bio-inspired synaptic plasticity.
- Navid Shervani-Tabar
- & Robert Rosenbaum
-
Article
| Open AccessA comprehensive benchmarking with practical guidelines for cellular deconvolution of spatial transcriptomics
This study comprehensively benchmarks 18 state-of-the-art methods for cellular deconvolution of spatial transcriptomics and provide decision-tree-style guidelines and recommendations for method selection.
- Haoyang Li
- , Juexiao Zhou
- & Xin Gao
-
Article
| Open AccessBenchmarking integration of single-cell differential expression
Integration of single-cell RNA sequencing data between different samples has been a major challenge for analyzing cell populations. Here the authors benchmark 46 workflows for differential expression analysis of single-cell data with multiple batches and suggest several high-performance methods under different conditions based on simulation and real data analyses.
- Hai C. T. Nguyen
- , Bukyung Baik
- & Dougu Nam
-
Article
| Open AccessAccelerating network layouts using graph neural networks
Visualization of large complex networks is challenging with limitations for the network size and depicting specific structures. The authors propose a Graph Neural Network based algorithm with improved speed and the quality of graph layouts, which allows for fast and informative visualization of large networks.
- Csaba Both
- , Nima Dehmamy
- & Albert-László Barabási
-
Article
| Open AccessFundamental limits to learning closed-form mathematical models from data
Learning analytical models from noisy data remains challenging and depends essentially on the noise level. The authors analyze the transition of the model-learning problem from a low-noise phase to a phase where noise is too high for the underlying model to be learned by any method, and estimate upper bounds for the transition noise.
- Oscar Fajardo-Fontiveros
- , Ignasi Reichardt
- & Roger Guimerà
-
Article
| Open AccessASGARD is A Single-cell Guided Pipeline to Aid Repurposing of Drugs
The full potential of single-cell RNA-sequencing applied to precision medicine has yet to be reached. Here, we propose a drug recommendation system ASGARD, which predicts drugs by considering cell clusters to address the intercellular heterogeneity within each patient.
- Bing He
- , Yao Xiao
- & Lana X. Garmire
-
Article
| Open AccessSearching for spin glass ground states through deep reinforcement learning
Finding the ground states of spin glasses relevant for disordered magnets and many other physical systems is computationally challenging. The authors propose here a deep reinforcement learning framework for calculating the ground states, which can be trained on small-scale spin glass instances and then applied to arbitrarily large ones.
- Changjun Fan
- , Mutian Shen
- & Yang-Yu Liu
-
Article
| Open AccessCartography of Genomic Interactions Enables Deep Analysis of Single-Cell Expression Data
Existing genomic data analysis methods tend to not take full advantage of underlying biological characteristics. Here, the authors leverage the inherent interactions of scRNA-seq data and develop a cartography strategy to contrive the data into a spatially configured genomap for accurate deep pattern discovery.
- Md Tauhidul Islam
- & Lei Xing
-
Article
| Open AccessLearning local equivariant representations for large-scale atomistic dynamics
The paper presents a method that allows scaling machine learning interatomic potentials to extremely large systems, while at the same time retaining the remarkable accuracy and learning efficiency of deep equivariant models. This is obtained with an E(3)- equivariant neural network architecture that combines the high accuracy of equivariant neural networks with the scalability of local methods.
- Albert Musaelian
- , Simon Batzner
- & Boris Kozinsky
-
Article
| Open AccessControllable branching of robust response patterns in nonlinear mechanical resonators
Feedback control applied to mechanical resonators can lead to the formation of various complex dynamic behaviors. Here the authors demonstrate flexible and controllable switching between dynamical structures in the response of harmonically driven micro-mechanical resonators.
- Axel M. Eriksson
- , Oriel Shoshani
- & David A. Czaplewski
-
Article
| Open AccessLeveraging molecular structure and bioactivity with chemical language models for de novo drug design
Generative Deep Learning holds promise for mining the unexplored “chemical universe” for new drugs. Here, the authors demonstrate the de novo design of phosphoinositide 3-kinase gamma (PI3Kγ) inhibitors for the PI3K/Akt pathway in human tumor cells.
- Michael Moret
- , Irene Pachon Angona
- & Gisbert Schneider
-
Article
| Open AccessPhysical deep learning with biologically inspired training method: gradient-free approach for physical hardware
Traditional learning procedures for artificial intelligence rely on digital methods not suitable for physical hardware. Here, Nakajima et al. demonstrate gradient-free physical deep learning by augmenting a biologically inspired algorithm, accelerating the computation speed on optoelectronic hardware.
- Mitsumasa Nakajima
- , Katsuma Inoue
- & Kohei Nakajima
-
Matters Arising
| Open AccessReply to: A balanced measure shows superior performance of pseudobulk methods in single-cell RNA-sequencing analysis
- Kip D. Zimmerman
- , Ciaran Evans
- & Carl D. Langefeld
-
Matters Arising
| Open AccessA balanced measure shows superior performance of pseudobulk methods in single-cell RNA-sequencing analysis
- Alan E. Murphy
- & Nathan G. Skene
-
Article
| Open AccessSeismic multi-hazard and impact estimation via causal inference from satellite imagery
This study presents the first rapid seismic multi-hazard and impact estimation system integrating advanced causal inference and remote sensing techniques, which jointly estimates regional-scale and high-resolution maps of seismic multi-hazards and building damage from InSAR imageries.
- Susu Xu
- , Joshua Dimasaka
- & Hae Young Noh
-
Article
| Open AccessNegotiation and honesty in artificial intelligence methods for the board game of Diplomacy
Artificial Intelligence has achieved success in a variety of single-player or competitive two-player games with no communication between players. Here, the authors propose an approach where Artificial Intelligence agents have ability to negotiate and form agreements, playing the board game Diplomacy.
- János Kramár
- , Tom Eccles
- & Yoram Bachrach
-
Article
| Open AccessPrecise atom manipulation through deep reinforcement learning
Engineering quantum states requires precise manipulations at the atomic level. Here, the authors use deep reinforcement learning to manipulate Ag adatoms on Ag surfaces, which combined with path planning algorithms enables autonomous atomic assembly.
- I-Ju Chen
- , Markus Aapro
- & Adam S. Foster
-
Article
| Open AccessDeciphering clinical abbreviations with a privacy protecting machine learning system
Patient notes contain shorthand and abbreviations that may be jargon or clinical vernacular. Here the authors train large machine learning models on public web data to decode such text by replacing abbreviations with their meanings.
- Alvin Rajkomar
- , Eric Loreaux
- & Juraj Gottweis
-
Article
| Open AccessAn AI approach for managing financial systemic risk via bank bailouts by taxpayers
Systemic risk and bank bailout approaches have been the source of discussions on scientific, financial and governmental forums. An artificial intelligence technique is proposed to inform equitable bailout decisions that minimise taxpayers’ losses.
- Daniele Petrone
- , Neofytos Rodosthenous
- & Vito Latora
-
Article
| Open AccessDeep autoencoder for interpretable tissue-adaptive deconvolution and cell-type-specific gene analysis
Traditional bulk sequencing data lack information about cell-type-specific gene expression. Here, the authors develop a Tissue-AdaPtive autoEncoder (TAPE), a deep learning method connecting bulk RNA-seq and single-cell RNA-seq, and apply it to analyze the cell type fractions and cell-type-specific gene expression in clinical data.
- Yanshuo Chen
- , Yixuan Wang
- & Yu Li
-
Article
| Open AccessUncertainty-informed deep learning models enable high-confidence predictions for digital histopathology
Safe clinical deployment of deep learning models for digital pathology requires reliable estimates of predictive uncertainty. Here the authors describe an algorithm for quantifying whole-slide image uncertainty, demonstrating their approach with models trained to distinguish lung cancer subtypes.
- James M. Dolezal
- , Andrew Srisuwananukorn
- & Alexander T. Pearson
-
Article
| Open AccessTechnology readiness levels for machine learning systems
The development of machine learning systems has to ensure their robustness and reliability. The authors introduce a framework that defines a principled process of machine learning system formation, from research to production, for various domains and data scenarios.
- Alexander Lavin
- , Ciarán M. Gilligan-Lee
- & Yarin Gal
-
Article
| Open AccessGenetic model of the El Laco magnetite-apatite deposits by extrusion of iron-rich melt
Can volcanoes erupt ore deposits? This study combines observations, experiments, and simulations to show that iron ore deposits on El Laco volcano formed by eruption of melt sourced from separation of Fe-rich melt from a silicate magma body beneath.
- Tobias Keller
- , Fernando Tornos
- & Jenny Suckale
-
Article
| Open AccessDiffraction-engineered holography: Beyond the depth representation limit of holographic displays
Improving the image depth perception of holograms while maintaining high image quality is a current challenge. Here the authors propose an efficient solution relying on a multi-plane hologram technique that reconstruct different blurred images and sharply focused images depending on a propagation distance.
- Daeho Yang
- , Wontaek Seo
- & Hong-Seok Lee
-
Article
| Open AccessAdversarial attacks and adversarial robustness in computational pathology
Artificial Intelligence can support diagnostic workflows in oncology, but they are vulnerable to adversarial attacks. Here, the authors show that convolutional neural networks are highly susceptible to white- and black-box adversarial attacks in clinically relevant classification tasks.
- Narmin Ghaffari Laleh
- , Daniel Truhn
- & Jakob Nikolas Kather
-
Article
| Open AccessUnderstanding Braess’ Paradox in power grids
Increasing the capacity of existing lines or adding new lines in power grids may, counterintuitively, reduce the system performance and promote blackouts. The authors propose an approach for prediction of edges that lower system performance and defining potential constrains for grid extensions.
- Benjamin Schäfer
- , Thiemo Pesch
- & Marc Timme
-
Article
| Open AccessDeep learning for twelve hour precipitation forecasts
Can AI learn from atmospheric data and improve weather forecasting? The neural network MetNet-2 achieves this by forecasting the fast changing variable of precipitation up to 12 h ahead more accurately and efficiently than traditional models based on hand-coded physics.
- Lasse Espeholt
- , Shreya Agrawal
- & Nal Kalchbrenner
-
Article
| Open AccessGeneralisable 3D printing error detection and correction via multi-head neural networks
3D printing is prone to errors and continuous monitoring and real-time correction during processing remains a significant challenge limiting its applied potential. Here, authors train a neural network to detect and correct diverse errors in real time across many geometries, materials and even printing setups.
- Douglas A. J. Brion
- & Sebastian W. Pattinson
-
Article
| Open AccessExplaining a series of models by propagating Shapley values
Series of machine learning models, relevant for tasks in biology, medicine, and finance, usually involve complex feature attribution techniques. The authors introduce a tractable method to compute local feature attributions for a series of machine learning models inspired by connections to the Shapley value.
- Hugh Chen
- , Scott M. Lundberg
- & Su-In Lee
-
Article
| Open AccessProtGPT2 is a deep unsupervised language model for protein design
Protein design aims to build novel proteins customized for specific purposes, thereby holding the potential to tackle many environmental and biomedical problems. Here the authors apply some of the latest advances in natural language processing, generative Transformers, to train ProtGPT2, a language model that explores unseen regions of the protein space while designing proteins with nature-like properties.
- Noelia Ferruz
- , Steffen Schmidt
- & Birte Höcker
-
Article
| Open AccessLead federated neuromorphic learning for wireless edge artificial intelligence
Designing energy-efficient computing solution for the implementation of AI algorithms in edge devices remains a challenge. Yang et al. proposes a decentralized brain-inspired computing method enabling multiple edge devices to collaboratively train a global model without a fixed central coordinator.
- Helin Yang
- , Kwok-Yan Lam
- & H. Vincent Poor
-
Article
| Open AccessStress-testing the resilience of the Austrian healthcare system using agent-based simulation
As mass quarantines, absences due to sickness, or other shocks thin out patient-physician networks, the system might be pushed to a tipping point where it loses its ability to deliver care. Here, the authors propose a data-driven framework to quantify regional resilience to such shocks via an agent-based model.
- Michaela Kaleta
- , Jana Lasser
- & Peter Klimek
-
Article
| Open AccessIdentifying multicellular spatiotemporal organization of cells with SpaceFlow
A critical task in spatial transcriptomics analysis is to understand inherently spatial relationships among cells. Here, the authors present a deep learning framework to integrate spatial and transcriptional information, spatially extending pseudotime and revealing spatiotemporal organization of cells.
- Honglei Ren
- , Benjamin L. Walker
- & Qing Nie
-
Article
| Open AccessAutomated detection and segmentation of non-small cell lung cancer computed tomography images
Correct interpretation of computer tomography (CT) scans is important for the correct assessment of a patient’s disease but can be subjective and timely. Here, the authors develop a system that can automatically segment the non-small cell lung cancer on CT images of patients and show in an in silico trial that the method was faster and more reproducible than clinicians.
- Sergey P. Primakov
- , Abdalla Ibrahim
- & Philippe Lambin
-
Article
| Open AccessAccelerating amorphous polymer electrolyte screening by learning to reduce errors in molecular dynamics simulated properties
Screening polymer electrolytes for batteries is extremely expensive due to the complex structures and slow dynamics. Here the authors develop a machine learning scheme to accelerate the screening and explore a space much larger than past studies.
- Tian Xie
- , Arthur France-Lanord
- & Jeffrey C. Grossman
-
Article
| Open AccessLearning emergent partial differential equations in a learned emergent space
Machine learning tools allow to extract dynamical systems from data, however this problem remains challenging for networks and systems of interacting agents. The authors introduce an approach to learn a predictive model for the dynamics of coupled agents in the form of partial differential equations using emergent spatial embeddings.
- Felix P. Kemeth
- , Tom Bertalan
- & Ioannis G. Kevrekidis
-
Article
| Open AccessDeciphering quantum fingerprints in electric conductance
Scattering of electrons from defects and boundaries in mesoscopic samples is encoded in quantum interference patterns of magneto-conductance, but these patterns are difficult to interpret. Here the authors use machine learning to reconstruct electron wavefunction intensities and sample geometry from magneto-conductance data.
- Shunsuke Daimon
- , Kakeru Tsunekawa
- & Eiji Saitoh
-
Article
| Open AccessMulti-scale turbulence simulation suggesting improvement of electron heated plasma confinement
Understanding the transport of the particles and fuel in the fusion plasma is fundamentally important. Here the authors report a cross-link interaction between electron- and ion-scale turbulences in plasma in terms of trapped electron mode and electron temperature gradient modes and their implication to fusion plasma.
- Shinya Maeyama
- , Tomo-Hiko Watanabe
- & Akihiro Ishizawa