Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
Single-cell datasets continue to grow in size and complexity, calling for computational tools to process and analyse data. Yang et al. present a contrastive learning framework to learn cell representations from single-cell multiomics datasets.
Federated learning and unsupervised anomaly detection are common techniques in machine learning. The authors combine them, using multicentred datasets and various diseases, to automate the segmentation of brain abnormalities without the need for annotations or sharing private local data.
Finding stable radical compounds for redox flow batteries is a challenging molecular design task. Sowndarya et al. combine an AlphaZero-based framework with a surrogate objective function trained on quantum chemistry simulations to generate suitable radical candidates that are stable. The approach promises to contribute to the development of low-cost, reliable energy storage technologies.
Targeted drug delivery is an exciting application of nanorobotics, but directing particles in the blood stream to the right location and in sufficient number is challenging. Gu and colleagues have developed a microtubule scaffold with embedded micromagnets that allows cargo, such as drug particles, to be transported in microvascular networks with precision and speed.
So-called noisy intermediate-scale quantum devices will be capable of a range of quantum simulation tasks, provided that the effects of noise can be sufficiently reduced. A neural error mitigation approach is developed that uses neural networks to improve the estimates of ground states and ground-state observables of molecules and quantum systems obtained using quantum simulations on near-term devices.
Using the natural dynamics of a legged robot for locomotion is challenging and can be computationally complex. A newly designed quadruped robot called Morti uses a central pattern generator inside two feedback loops as an adaptive method so that it efficiently uses the passive elasticity of its legs and can learn to walk within 1 h.
Artificial DNA circuits that can perform neural network-like computations have been developed, but scaling up these circuits to recognize a large number of patterns is a challenging task. Xiong, Zhu and colleagues experimentally demonstrate a convolutional neural network algorithm using a synthetic DNA-based regulatory circuit in vitro and develop a freeze–thaw approach to reduce the computation time from hours to minutes, paving the way towards more powerful biomolecular classifiers.
An end-to-end machine learning approach that can learn which mechanisms determine cell fate and competition from a large time-lapse microscopy dataset is developed. The approach makes use of a probabilistic autoencoder to learn an interpretable representation of the organization of cells, and provides cell fate predictions that can be tested in drug screening experiments.
Deep learning methods can provide useful predictions for drug design, but their hyperparameters need to be carefully tweaked to give good performance on a specific problem or dataset. Li et al. present here a method that finds appropriate architectures and hyperparameters for a wide range of drug design tasks and can achieve good performance without human intervention.
Exoskeletons can assist movement in upper limb impairments to recover mobility and independence, but rigid or heavy exoskeletons can be impractical. Georgarakis and colleagues have developed a soft, tendon-driven device that assists shoulder movements and counteracts gravity to reduce muscle fatigue.
Neural networks in the brain often exhibit chaotic dynamics that can be captured by a small number of dimensions. Farrell et al. find that recurrent neural networks trained with gradient-based learning rules exhibit similar features. This helps form robust but generalizable input representations.
Robots usually learn to use tools from direct experience or from observing the use of a tool. While knowledge can be transferred between similar tools, novel and creative use of tools is challenging. Tee and colleagues present a method where skill transfer does not come from experience of using other tools but from using the robot’s own limbs.
While reinforcement learning can be a powerful tool for complex design tasks such as molecular design, training can be slow when problems are either too hard or too easy, as little is learned in these cases. Jeff Guo and colleagues provide a curriculum learning extension to the REINVENT de novo molecular design framework that provides problems of increasing difficulty over epochs such that the training process is more efficient.
A variational autoencoder is trained on a dataset of quantum optics experiment configurations and learns an interpretable representation of the relationship between experiment setup and quantum entanglement. The approach can be used to explore new experiment designs with specific, highly entangled states.
B-cell receptors (BCRs) and their impact on B cells play a vital role in our immune system; however, the manner in which B cells are activated by BCRs are still poorly understood. Ze Zhang and colleagues present a graph-based method that connects BCR and single B-cell RNA sequencing data and identifies notable coupling between BCR and B-cell expression in COVID-19.
Fluorescent markers in microscopy-based drug screens carry information on how compounds affect biological processes, but practical considerations may hinder their use. Wong et al. develop a deep learning method for generating images in drug discovery, with broad applicability across diverse fluorescence microscopy datasets.
Tactile sensing is needed for robots to physically interact with humans in daily living and in the workplace. A scientific challenge in robotics is how to simultaneously detect contact location and intensity. The authors describe a large-area sensing skin for robotic system applications, specifically for human–machine interactions.
Respiratory complications after a COVID infection are a growing concern, but follow-up chest CT scans of COVID-19 survivors hardly present any recognizable lesions. A deep learning-based method was developed that calculates a scan-specific optimal window and removes irrelevant tissues such as airways and blood vessels from images with segmentation models, so that subvisual abnormalities in lung scans become visible.
Deep learning could be less energy intensive when implemented on spike-based neuromorphic chips. An approach inspired by a characteristic feature of biological neurons, the presence of slowly changing internal currents, is developed to emulate long short-term memory units in a sparse spiking regime for neuromorphic implementation.
Swarms of microrobots could eventually be used to deliver drugs to specific targets in the body, but the coordination of these swarms in complex environments is challenging. Yang and colleagues present a real-time autonomous distribution planning method based on deep learning that controls both the shape and position of the swarm, as well as the imaging system used for swarm navigation to cover longer distances.