Articles in 2020

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.

Robot-assisted microsurgery promises high stability and accuracy for instance in eye- or neurosurgery applications. A new miniature robotics device, based on an origami-inspired design, can make complex 3D motions and reaches a precision of around 26 micrometres.

Machine learning has become popular in solving complex optical problems such as recovering the input phase and amplitude for a specific pattern or image measured through a scattering medium. In a more challenging application, Rahmani et al. consider the problem of also producing desired outputs for such a nonlinear system when only some intensity-only measurements of example outputs are available. They develop a neural network approach that can ensure the transmission of images through a highly nonlinear system—a multimode fibre—with a 90% fidelity.

Deep learning methods can be a powerful part of digital pathology workflows, provided well-annotated training datasets are available. Tolkach and colleagues develop a deep learning model to recognize and grade prostate cancer, based on a convolution neural network and a dataset with high-quality labels at gland-level precision.

Currently available quantum hardware is limited by noise, so practical implementations often involve a combination with classical approaches. Sels et al. identify a promising application for such a quantum–classic hybrid approach, namely inferring molecular structure from NMR spectra, by employing a range of machine learning tools in combination with a quantum simulator.

A goal of biology is to identify the molecular mechanisms that control differential gene expression. Tasaki et al. have developed a framework that integrates genomic data into a deep learning model of transcriptome regulations to predict multiple transcriptional effects in tissue- and person-specific transcriptomes.

Tumour mutational burden (TMB) shows promise as a biomarker in cancer immunotherapy, but it usually requires whole-exome sequencing, which is costly, time-consuming and unavailable at most hospitals. The authors develop a machine learning algorithm that uses standard H&E histopathological images to quickly, inexpensively and accurately predict TMB. The approach may have applications as a tool to screen and prioritize patient samples and subsequent treatments.

Vascular abnormalities are challenging for diagnostic imaging due to the complexity of vasculature and the non-uniform scattering from biological tissues. The authors present an unsupervised learning algorithm for vascular feature recognition from small sets of biomedical images acquired from different modalities. They demonstrate the utility of their diagnostic approach on vascular images of thrombosis, internal bleeding and colitis.

Gene sets can provide valuable information for gaining insight into disease mechanisms and cellular functions. In this paper, the authors use a Gaussian approach to represent gene sets and gene networks in a low-dimensional space, allowing for accurate prediction and decreased computational complexity.

Spiking neural networks and in-memory computing are both promising routes towards energy-efficient hardware for deep learning. Woźniak et al. incorporate the biologically inspired dynamics of spiking neurons into conventional recurrent neural network units and in-memory computing, and show how this allows for accurate and energy-efficient deep learning.

A lot of scientific literature is unstructured, which makes extracting information for biomedical databases difficult. Hong and colleagues show that a distant supervision approach, using latent tree learning and recurrent units, can extract drug–target interactions from literature that were previously unknown.

A fundamental problem in network science is how to find an optimal set of key players whose activation or removal significantly impacts network functionality. The authors propose a deep reinforcement learning framework that can be trained on small networks to understand the organizing principles of complex networked systems.

The rise of deep neural networks allows for new ways to design molecules that interact with biological structures. An approach that uses conditional recurrent neural networks generates molecules with properties near specified conditions.