Articles

Microrobots are usually too small to contain traditional computing substrates that could control their behaviour. Dekanovsky and colleagues have developed a microrobot swarm that removes hormonal pollutants when it senses a chemical signal in its environment.

Metal–organic frameworks (MOFs) are attractive materials for gas capture, separation, sensing and catalysis. Determining their water stability is important, but time-intensive. Batra et al. use machine learning to screen water-stable MOFs and identify chemical features supporting their stability.

The thickness of the retina is an important medical indicator for diabetic retinopathy. Holmberg and colleagues present a self-supervised deep-learning method that uses cross-modal data to predict retinal thickness maps from easily obtainable fundus images.

The wealth of data gathered from single-cell RNA sequencing can be processed with deep learning techniques, but often those methods are too opaque to reveal why a single cell is labelled to be a certain cell type. Lifei Wang and colleagues present an RNA-sequencing analysis method that uses capsule networks and is interpretable enough to allow for identification of cell-type-specific genes.

Across disciplines, there is a rising interest in interpreting machine learning models to derive scientific knowledge from data. Genkin and Engel show that models optimized for predicting data can disagree with the ground truth and propose a new model selection principle to prioritize accurate interpretation.

Neural network models can predict the socioeconomic wealth of an area from aerial views, but fall short of explaining how visual features trigger a given prediction. The authors develop a pipeline for projecting class activation maps onto the underlying urban topology, to help interpret such predictions.

To infer a previously unknown molecular formula from mass spectrometry data is a challenging, yet neglected problem. Ludwig and colleagues present a network-based approach to ranking possible formulas.

Inspired by the brain of the roundworm Caenorhabditis elegans, the authors design a highly compact neural network controller directly from raw input pixels. Compared with larger networks, this compact controller demonstrates improved generalization, robustness and interpretability on a lane-keeping task.

Magnetic endoscopes have the potential to improve access, reduce patient discomfort and enhance safety. While navigation of magnetic endoscopes can be challenging for the operator, a new approach by Martin, Scaglioni and colleagues explores how to reduce this burden by offering different levels of autonomy in robotic colonoscopy.

Recent advances have increased the dimensionality and complexity of immunological data. The authors developed a machine learning approach to incorporate prior immunological knowledge and applied it on clinical examples and a simulation study. The approach may be useful for high-dimensional datasets in clinical settings where the cohort size is limited.

Classifying cells from single-cell RNA sequences is challenging for cells where only limited data is available. Hu and colleagues show here that a clustering approach using transfer learning can use the variability of one dataset to cluster a smaller target dataset with high-quality results.

Advances in large-scale connectivity mapping of the brain require efficient computational tools to detect fine structures across large volumes of images, which poses challenges. The authors introduce a hybrid architecture that incorporates topological priors of neuronal structures with deep learning models to improve semantic segmentation of neuroanatomical image data.

Recent accidents with autonomous test vehicles have eroded trust in such self-driving cars. A shift in approach is required to ensure autonomous vehicles can never be the cause of accidents. An online verification technique is presented that guarantees provably safe motions, including fallback solutions in safety-critical situations, for any intended trajectory calculated by the underlying motion planner.

Reinforcement learning has become a popular method in various domains, for problems where an agent must learn what actions must be taken to reach a particular goal. An interesting example where the technique can be applied is simulated annealing in condensed matter physics, where a procedure is determined for slowly cooling a complex system to its ground state. A reinforcement learning approach has been developed that can learn a temperature scheduling protocol to find the ground state of spin glasses, magnetic systems with strong spin–spin interactions between neighbouring atoms.

Training machine learning models to predict the function of proteins is limited by the availability of only a small amount of labelled training data. Training can be improved by employing generative adversarial networks to generate additional synthetic protein samples.

When automated decisions are provided by a company without providing the full model, users and law makers might demand a ‘right to an explanation’. Le Merrer and Trédan show that malicious manipulations of these explanations are hard to detect, even for simple strategies to obscure the model’s decisions.

When designing new drugs, there are countless ways to create molecules, yet only a few interact with biological targets. Beker and colleagues provide here a graph neural network based metric for drug-likeness that can guide the search.

Deep learning approaches can show excellent performance but still have limited practical use if they learn to predict based on confounding factors in a dataset, for instance text labels in the corner of images. By using an explanatory interactive learning approach, with a human expert in the loop during training, it becomes possible to avoid predictions based on confounding factors.

The role of DNA methylation on N⁶-adenine (6mA) in eukaryotes is a challenging research problem. Tan et al. develop a deep-learning-based algorithm to predict 6mA sites from sequences at single-nucleotide resolution, and apply the method to three representative model organisms. The method is further developed to visualize regulatory patterns around 6mA sites.

Gene expression is regulated by a variety of mechanisms, which have been difficult to study in a unified way. The authors propose a flexible framework that can integrate different types of data for studying their joint effects on gene expression. The framework uses a general network representation for data integration, metapaths for inputting prior knowledge of gene regulatory mechanisms, and embedding techniques for capturing complex structures in the data.