Research articles

Predicting the properties of a molecule from its structure with high accuracy is a crucial problem in digital drug design. Instead of sequence features, Zeng and colleagues use an image representation of a large collection of bioactive molecules to pretrain a model that can be fine-tuned on specific property prediction tasks.

Predicting disease progression is an important medical problem, but it can be challenging for end-to-end machine learning approaches. Han and colleagues demonstrate that generative models can work together with medical experts to jointly predict the progression of a disease, osteoarthritis.

Physical dynamical processes can be modelled with differential equations that may be solved with numerical approaches, but this is computationally costly as the processes grow in complexity. In a new approach, dynamical processes are modelled with closed-form continuous-depth artificial neural networks. Improved efficiency in training and inference is demonstrated on various sequence modelling tasks including human action recognition and steering in autonomous driving.

Computational methods are important for interpreting missense variants in genetic studies and clinical testing. Zhang and colleagues develop a method based on graph attention neural networks to predict pathogenic missense variants. The method pools information from functionally correlated positions and can improve the interpretation of missense variants.

To enable electrification of cities, thereby decarbonizing energy grids, it is essential to predict electricity consumption with high spatio-temporal accuracy. To reduce the need for large amounts of training data, an active deep learning approach is developed to make accurate forecasts of electric load profiles at the scale of single buildings.

The effects of novel antibodies are hard to predict owing to the complex interactions between antibodies and antigens. Zhang and colleagues use a graph-based method to learn a dynamic representation that allows for predictions of neutralization activity and demonstrate the method by recommending probable antibodies for human immunodeficiency virus, severe acute respiratory syndrome coronavirus 2, influenza and dengue.

To be useful in human life, robots need to learn the social rules of human society. Zhou et al. investigate the social rules that apply in spaces mutually occupied by humans and robots. The authors develop a social locomotion model for a mobile robot and implement it for socially aware navigation.

Image noise is a common problem in light microscopy, and denoising is a key step in microscopic imaging pipelines. Lequyer et al. propose a self-supervised denoising method and apply it to diverse imaging and analysis pipelines.

Single-cell multi-omics technologies have increased dramatically in biomedical research. Lakkis et al. develop a deep learning method to address computational challenges in CITE-seq and single-cell RNA-seq datasets.

Recognition of speech from lip movements is still a challenging problem and much effort is concentrated on the English language. Ma et al. have used auxiliary tasks to train a model such that it works for a range of different languages, including Mandarin, Spanish, Italian, French and Portuguese.

Identifying epidemic hotspots in a timely way with syndromic surveillance can provide highly valuable information for public health policy. A machine learning early indicator model that uses highly granular data from digitalized healthcare-seeking behaviour, including from Google Trends and National Health Service Pathways calls, can identify SARS-CoV-2 risk at small geographic scales. The model can retrospectively identify hotspots in the United Kingdom for various variants in 2020 and 2021 before the wider spread and growth of these variants being confirmed by clinical data.

The haptic interface is an essential part of human–machine interfaces where tactile information is delivered between human and machine. Yao et al. develop a soft, ultrathin, miniaturized and wireless electrotactile system that allows virtual tactile information to be reproduced over the hand.

Predicting patient-specific clinical drug responses from cell-line screens using machine learning is challenging. He and colleagues develop a deep learning method to predict patient-specific clinical responses from cell-line and other disease models for drug discovery and personalized medicine.

Saliency methods are used to localize areas of medical images that influence machine learning model predictions, but their accuracy and reliability require investigation. Saporta and colleagues evaluate seven saliency methods using different model architectures, and find that saliency maps perform worse than a human radiologist benchmark.

Earth system models (ESMs) are powerful tools for simulating climate fields, but weather forecasting and in particular precipitation prediction with ESMs are challenging. A generative adversarial network, constrained by the sum of global precipitation, is developed that substantially improves ESM predictions of spatial patterns and intermittency of daily precipitation.

Cell type annotation is a core task for single cell RNA-sequencing, but current bioinformatic tools struggle with some of the underlying challenges, including high dimensionality, data sparsity, batch effects and a lack of labels. In a self-supervised approach, a transformer model called scBERT is pretrained on millions of unlabelled public single cell RNA-seq data and then fine-tuned with a small number of labelled samples for cell annotation tasks.

The space of possible proteins is vast, and optimizing proteins for specific target properties computationally is an ongoing challenge, even with large amounts of data. Castro and colleagues combine a transformer-based model with regularized prediction heads to form a smooth and pseudoconvex latent space that allows for easier navigation and more efficient optimization of proteins.

The performance of machine learning models is usually compared via the mean value of a selected performance measure such as the area under the receiver operating characteristic curve on a specific benchmark data set. However, this measure, its mean value or even relative differences between models do not provide a good prediction of whether the results can translate to other data sets. Gosiewska and colleagues present here a comparison based on Elo ranking, which offers a probabilistic interpretation of how much better one model is than another.

Producing high-quality 3D refractive index maps from 2D intensity-only measurements is a long-standing objective in computational microscopy, with many applications involving the visualization of cellular and subcellular structures. A new method can reconstruct high-contrast and artefact-free images by employing the neural fields technique, which can learn a continuous 3D representation using a neural network that maps spatial coordinates to the refractive index values.