Signals, in order.

This research demonstrates how machine learning combined with volcano plot analysis can identify new cross‑coupling catalysts. The study was conducted in Switzerland and focuses on computational methods to accelerate catalyst discovery.

Source: opendata.swiss

ETH Zurich researchers have shown how the two membranes of the nuclear envelope fuse to form nuclear pore complexes, a process requiring the enzyme Torsin-1A and a complex of CLCC1 proteins. The study explains how defects in Torsin-1A halt pore assembly and links this mechanism to the hereditary disease DYT1 dystonia, suggesting a potential therapeutic approach by boosting CLCC1 levels. The findings were published in Molecular Cell and involve collaboration with the University of Fribourg, PSI, EMBL, and RWTH Aachen University.

Source: ETH Zurich News

EPFL has released an open‑source CO₂ calculator that provides research laboratories with a detailed, granular view of their carbon emissions, allowing scientists to identify the most impactful reduction measures. The tool automatically imports data such as lab surface area, purchasing history and power use, and has already been adopted by EPFL labs and is being explored by other Swiss universities, including the University of Lausanne and ETH Zurich.

Source: EPFL News

The paper proposes a latent style-based quantum Wasserstein GAN (QGAN) for drug design, combining a variational autoencoder that maps molecular structures into a latent space with a hybrid quantum GAN. It benchmarks against classical MOSES suite models, achieving comparable performance with far fewer trainable parameters (110 versus 700,000). Experiments include training on noiseless quantum simulators with up to 15 qubits and an inference-only demonstration on a 156‑qubit IBM Heron quantum computer using a five‑qubit setup, sampling 2,500 new molecules to study quantum noise effects.

Source: OpenAlex

Eva Lieberherr argues that Switzerland’s current centralized water system is inefficient and unsustainable, especially under climate change pressures. She proposes a decentralized, modular, and circular approach that treats and reuses water locally, recovering nutrients for fertilizer. Pilot projects, such as in Geneva, demonstrate feasibility, but only 2.5 % of wastewater is currently handled this way, indicating a need for more pilots and potential legal adjustments.

Source: ETH Zurich News

Researchers at EPFL discovered that Pseudomonas aeruginosa bacteria sense physical collisions and reverse direction within seconds, allowing them to avoid becoming trapped in dense groups. Wild‑type bacteria use this mechanosensing to maintain mobility in crowded areas while aligning at colony edges, enabling efficient exploration of maze‑like environments. The study, published in Nature Microbiology, demonstrates active regulation of collective behavior through physical feedback rather than chemical signals.

Source: EPFL News

The Paul Scherrer Institute (PSI) in Villigen is extending optogenetics beyond nerve cells by developing light-sensitive proteins that can control signalling pathways in various cell types. A team led by Gebhard Schertler is creating optoGPCRs to manipulate G‑protein‑coupled receptors, potentially enabling new treatments for cardiac arrhythmias and Parkinson’s disease. Time‑resolved crystallography at SwissFEL and the Swiss Light Source (SLS) is used to observe the rapid structural changes of these proteins after light activation.

Source: Swiss Federal Government

The paper argues that human‑AI complementarity, the idea that a human supported by AI can outperform either alone, faces theoretical challenges such as lack of precise anchoring and reliance on post‑hoc accuracy. By applying epistemology and computational reliabilism, the authors reframe complementarity as evidence of a reliable epistemic process, linking it to other reliability indicators and socio‑technical practices. They propose design and governance recommendations, including a minimal reporting checklist and measures of efficient complementarity.

Source: OpenAlex

Large language models (LLMs) are increasingly used for collective reasoning on complex, value-laden problems, yet their performance on non‑verifiable, pluralistic tasks cannot be fully inferred from standard benchmarks. Using the Deliberative Reason Index (DRI), the study evaluated 1,980 five‑agent LLM runs across 12 citizen‑assembly topics and found that LLM groups produce procedural discourse comparable to humans but show limited gains in intersubjective consistency, reduced perspective diversity, and a reversal of human convergence patterns. Engineering diversity through persona prompting …

Source: OpenAlex

This chapter explores the ergonomic, cognitive, and psychological mechanisms underpinning human–robot communication, focusing on how social robots shape human perception, behaviour, and socio-emotional engagement. It examines robot design, embodiment, and communication modalities, addressing reciprocity and behavioural appropriateness, and discusses the implications of adaptive technologies such as large language models for responsiveness and confidentiality. The chapter also highlights the potential of social robots in workplace and therapeutic settings to enhance efficiency, safety, and emo…

Source: OpenAlex