Reviews & Analysis

We created an open-source model that simulates Caenorhabditis elegans in a closed-loop system, by integrating simulations of its brain, its physical body, and its environment. BAAIWorm replicated C. elegans locomotive behaviors, and synthetic perturbations of synaptic connections impacted neural control of movement and affected the embodied motor behavior.

Inspired by recent approaches for natural language processing and computer vision, we developed Annotatability, a framework that analyzes deep neural network training dynamics to interpret pre-annotated single-cell and spatial omics data. Annotatability identified erroneous annotations and ambiguous cell states, inferred trajectories from binary labels, and revealed underlying biological signals.

By developing an efficient spin symmetry penalty, a recent study has substantially accelerated the calculation of accurate energies with correct spin states in variational Monte Carlo for both ground and excited states of quantum many-particle systems.

A recent study proposes DeepBlock, a deep learning-based approach for generating ligands with targeted properties, such as low toxicity and high affinity with the given target. This approach outperforms existing methods in the field while maintaining synthetic accessibility and drug-likeness.

A recent study has modeled and quantified the expected rise in electronic waste due to the increasing deployment of generative artificial intelligence.

A recent study introduces a series of approaches that predict protein fitness and stability after the introduction of mutations. The work focuses on combining different data and pre-training to overcome data scarcity.

As digital data expand exponentially, traditional storage media are becoming less viable, making DNA a promising solution due to its density and durability. In this Perspective, the authors discuss the critical computational challenges associated with in vitro DNA-based data storage.

A method is introduced to compute provable bounds on noise-free quantum expectation values from noisy samples, promising potential applications in quantum optimization and machine learning.

Active machine learning is employed in academia and industry to support drug discovery. A recent study unravels the factors that influence a deep learning models’ ability to guide iterative discovery.

The application of machine learning techniques to small-molecule drug discovery has not yet yielded a true leap forward in the field. This Perspective discusses how a renewed focus on data and validation could help unlock machine learning’s potential.

A recent study proposes a strategy for the prediction of genetic perturbation outcomes by breaking it down into three subtasks: identifying differentially expressed genes, determining expression change directions, and estimating gene expression magnitudes.

A deep learning algorithm is presented to classify single-particle tracking trajectories into theoretical models of anomalous diffusion and detect if the trajectory is related to a model not originally found within the training dataset.

A highly efficient reconstruction method has been developed for the direct computation of Hamiltonian matrices in the atomic orbital basis from density functional theory calculations originally performed in the plane wave basis. This enables machine learning calculations of electronic structures on a large scale, which are otherwise not feasible with standard methods, and thus fills a methodological gap in terms of accessible length scales.

Nature Computational Science asked a group of scientists to discuss strategies for increasing the presence of Black, Indigenous, People of Color (BIPOC) researchers in computational science, as well as the various considerations to be made for improving education and methods design.

We developed mixture model inference with discrete-coupled autoencoders (MMIDAS), an unsupervised variational framework that jointly learns discrete clusters and continuous cluster-specific variability. When applied to unimodal or multimodal single-cell omic data, MMIDAS learned single-cell representations with robust cell type definitions and interpretable, continuous within-cell type variability.

A recent study proposes an approach that integrates unspliced and spliced mRNA count data by leveraging shared biophysical states across cells, offering a more interpretable and consistent framework for determining cell clusters based on transcriptional kinetics.

A computational model is proposed to provide a better understanding of human altruism, highlighting the role of multiple motives that influence altruistic behaviors.

Mechanical metamaterials have shown potential for processing information via autonomous environmental interactions. This Perspective summarizes recent efforts and challenges on integrating stimuli-responsive materials with mechanical metamaterials for mechanical computing, and explores the remaining challenges in the field.

Computational tools have recently empowered mechanical metamaterials design. In this Perspective, advances to these approaches are discussed, notably mechanism-based design, topology optimization, the use of machine learning and the challenges for additive-manufactured metamaterial structures.

Revealing a drug’s mechanism of action (MOA) is costly and time-consuming. In this study, we used deep learning to extract temporal mitochondrial phenotypic features after exposure to drugs with known MOAs using re-identification algorithms. The trained model could then predict the MOAs of unidentified substances, facilitating phenotypic screening-based drug discovery and repurposing.