Machine learning articles from across Nature Portfolio

A randomized phase 2a clinical trial of an AI-discovered drug and target combination for idiopathic pulmonary fibrosis shows safety and signs of efficacy, marking a concrete step forward in bringing AI-enabled drug discovery into the clinic.

Technologies that can interpret imagery have many potential applications. An innovative study of papers and patents in the field suggests one use case that overrides the others: human surveillance.

An accessible and adaptable artificial-intelligence model trained on a diverse set of X-ray images is better than existing models at spotting rare chest diseases.

Comparative Physicochemical Profiling (CPP), an interpretable machine learning algorithm, identifies physicochemical signatures of γ-secretase substrates, capturing features hidden in sequences and enabling proteome-wide prediction of substrate recognition.

Single-cell CRISPR screens enable high-throughput analysis of how genetic changes affect individual cells. Here, authors present GPerturb, a method that accurately detects and quantifies gene-level effects of perturbations, with uncertainty estimates, revealing complex biological interactions.

Cascadia is a mass spectrometry-based de novo sequencing model that uses a transformer architecture to handle data-independent acquisition data and achieves substantially improved performance across a range of instruments and experimental protocols.

An associative-learning-enhanced model, called OrganADR, is developed to predict adverse drug reactions at the organ level. OrganADR could be applied in emerging combination pharmacotherapies and improve medication safety.

ColdBrew is a machine learning-based method that predicts the probability of cryogenic crystallographic waters to be present at room temperature, which links to their relative energies and displaceability by ligands.

A randomized phase 2a clinical trial of an AI-discovered drug and target combination for idiopathic pulmonary fibrosis shows safety and signs of efficacy, marking a concrete step forward in bringing AI-enabled drug discovery into the clinic.

Several groups hope to develop artificial-intelligence models that can predict how cells behave.

The majority of papers and patents in a subfield of AI enable surveillance, according to a new study.

The output of computer-vision research is overwhelmingly aimed towards monitoring humans. The potential ethical implications need more scrutiny.

When researchers cede their scientific judgement to machines, we lose something important.