Explore the vast range of titles available.

Modern chemical manufacturing, on which human quality of life depends, is unsustainable; alternative production routes must be developed. Electrochemical and biological processes offer promise for upgrading waste streams, including recalcitrant carbon dioxide and plastic-derived wastes. However, the inherent heterogeneity and high energy requirements of upcycling the chemical endpoints of the “take-make-waste\" economy remain challenging. Cupriavidus necator is an emergent catalyst for complex feedstock valorization because of its extreme metabolic flexibility, which allows it to utilize a wide array of substrates, and its ability to use carbon dioxide via the Calvin-Benson-Bassham cycle. C. necator natively oxidizes hydrogen to power carbon utilization, but its flexibility offers an as-yet unexplored opportunity to couple waste stream oxidation with carbon dioxide utilization instead, potentially enabling carbon conservative waste upcycling. Here, we uncover the constraints on carbon conservative chemical transformation using C. necator as a model. We systematically examine the carbon yield and thermodynamic feasibility of mixotrophic scenarios combining waste-derived carbon sources with hydrogen oxidation to power carbon reassimilation. Then, we evaluate carbon-carbon mixotrophic scenarios, with one carbon source providing electrons in place of hydrogen oxidation. We show that both hydrogen and ethylene glycol are feasible electron sources to drive carbon-neutral or carbon-negative mixotrophic upgrading of waste streams such as acetate or butyrate. In contrast, we find that carbon conservation is likely infeasible for most other waste-derived carbon sources. This work provides a roadmap to establishing novel C. necator strains capable of carbon efficient waste upcycling.

Biomanufacturing offers a potentially sustainable alternative to deriving chemicals from fossil fuels. However, traditional biomanufacturing, which uses sugars as feedstocks, competes with food production and yields unfavourable land use changes, so more sustainable options are necessary. Cupriavidus necator is a chemolithoautotrophic bacterium capable of consuming carbon dioxide and hydrogen as sole carbon and energy sources, or formate as the source of both. This autotrophic metabolism potentially makes chemical production using C. necator sustainable and attractive for biomanufacturing. Additionally, C. necator natively fixes carbon in the form of poly-3-hydroxybutyrate, which can be processed to make biodegradable plastic. Recent progress in development of modelling and synthetic biology tools have made C. necator much more usable as a biomanufacturing chassis. However, these tools and applications are often limited by a lack of consideration for the unique physiology and metabolic features of C. necator . As such, further work is required to better understand the intricate mechanisms that allow it to prioritise generalization over specialization. In this review, progress toward physiology-informed engineering of C. necator across several dimensions is critically discussed, and recommendations for moving toward a physiological approach are presented. Arguments for metabolic specialization, more focus on autotrophic fermentation, C. necator -specific synthetic biology tools, and modelling that goes beyond constraints are presented based on analysis of existing literature.

Many nanoscale devices require precise optimization to function. Tuning them to the desired operation regime becomes increasingly difficult and time-consuming when the number of terminals and couplings grows. Imperfections and device-to-device variations hinder optimization that uses physics-based models. Deep neural networks (DNNs) can model various complex physical phenomena but, so far, are mainly used as predictive tools. Here, we propose a generic deep-learning approach to efficiently optimize complex, multi-terminal nanoelectronic devices for desired functionality. We demonstrate our approach for realizing functionality in a disordered network of dopant atoms in silicon. We model the input–output characteristics of the device with a DNN, and subsequently optimize control parameters in the DNN model through gradient descent to realize various classification tasks. When the corresponding control settings are applied to the physical device, the resulting functionality is as predicted by the DNN model. We expect our approach to contribute to fast, in situ optimization of complex (quantum) nanoelectronic devices.Function implementation and optimization in nanoscale and quantum-electronic devices become increasingly challenging with the growing complexity of the devices. Training a deep neural network with the physical device response and searching for the functionality in the digital device can ease this challenge.

Classification is an important task at which both biological and artificial neural networks excel 1 , 2 . In machine learning, nonlinear projection into a high-dimensional feature space can make data linearly separable 3 , 4 , simplifying the classification of complex features. Such nonlinear projections are computationally expensive in conventional computers. A promising approach is to exploit physical materials systems that perform this nonlinear projection intrinsically, because of their high computational density 5 , inherent parallelism and energy efficiency 6 , 7 . However, existing approaches either rely on the systems’ time dynamics, which requires sequential data processing and therefore hinders parallel computation 5 , 6 , 8 , or employ large materials systems that are difficult to scale up 7 . Here we use a parallel, nanoscale approach inspired by filters in the brain 1 and artificial neural networks 2 to perform nonlinear classification and feature extraction. We exploit the nonlinearity of hopping conduction 9 , 10 – 11 through an electrically tunable network of boron dopant atoms in silicon, reconfiguring the network through artificial evolution to realize different computational functions. We first solve the canonical two-input binary classification problem, realizing all Boolean logic gates 12 up to room temperature, demonstrating nonlinear classification with the nanomaterial system. We then evolve our dopant network to realize feature filters 2 that can perform four-input binary classification on the Modified National Institute of Standards and Technology handwritten digit database. Implementation of our material-based filters substantially improves the classification accuracy over that of a linear classifier directly applied to the original data 13 . Our results establish a paradigm of silicon-based electronics for small-footprint and energy-efficient computation 14 . The nonlinearity of hopping conduction in a disordered network of boron dopant atoms in silicon is used to perform nonlinear classification and feature extraction.

Deep learning, a multilayered neural-network approach inspired by the brain, has revolutionized machine learning. Its success relies on backpropagation, which computes gradients of a loss function for use in gradient descent. However, digital implementations are energy hungry, with power demands limiting many applications. This has motivated specialized hardware, from neuromorphic CMOS and photonic tensor cores to unconventional material-based systems. Learning in such systems, for example via artificial evolution, equilibrium propagation, or surrogate modelling, is typically complicated and slow. Here, we demonstrate a simple gradient-extraction method based on homodyne detection, enabling gradient descent directly in physical systems without the need for an analytical description. By perturbing parameters with sinusoidal waveforms at distinct frequencies, we robustly obtain gradient information in a scalable manner. We illustrate the method in reconfigurable nonlinear-processing units and argue for broad applicability. Homodyne gradient extraction can in principle be fully implemented in materia, facilitating autonomously learning material systems. Training deep neural networks by backpropagation consumes significant energy in digital hardware. Boon and Cassola et al. show that homodyne detection can be used to extract gradients directly in a physical device, enabling efficient gradient descent and offering a scalable route to material-based learning.

Adaptive laboratory evolution experiments provide a controlled context in which the dynamics of selection and adaptation can be followed in real-time at the single-nucleotide level. And yet this precision introduces hundreds of degrees-of-freedom as genetic changes accrue in parallel lineages over generations. On short timescales, physiological constraints have been leveraged to provide a coarse-grained view of bacterial gene expression characterized by a small set of phenomenological parameters. Here, we ask whether this same framework, operating at a level between genotype and fitness, informs physiological changes that occur on evolutionary timescales. Using a strain adapted to growth in glucose minimal medium, we find that the proteome is substantially remodeled over 40 000 generations. The most striking change is an apparent increase in enzyme efficiency, particularly in the enzymes of lower-glycolysis. We propose that deletion of metabolic flux-sensing regulation early in the adaptation results in increased enzyme saturation and can account for the observed proteome remodeling. Adaptive laboratory evolution provides a real-time record of physiological change. In bacteria adapted to glucose over 40 000 generations, this study finds an apparent increase in enzyme efficiency consistent with increased substrate saturation due to loss of a flux sensing mechanism early in adaptation.

Major airstreams in tropical cyclones (TCs) are rarely described from a Lagrangian perspective. Such a perspective, however, is required to account for asymmetries and time dependence of the TC circulation. We present a procedure that identifies main airstreams in TCs based on trajectory clustering. The procedure takes into account the TC’s large degree of inherent symmetry and is suitable for a very large number of trajectories . A large number of trajectories may be needed to resolve both the TC’s inner-core convection as well as the larger-scale environment. We define similarity of trajectories based on their shape in a storm-relative reference frame, rather than on proximity in physical space, and use Fréchet distance, which emphasizes differences in trajectory shape, as a similarity metric. To make feasible the use of this elaborate metric, data compression is introduced that approximates the shape of trajectories in an optimal sense. To make clustering of large numbers of trajectories computationally feasible, we reduce dimensionality in distance space by so-called landmark multidimensional scaling. Finally, k -means clustering is performed in this low-dimensional space. We investigate the extratropical transition of Tropical Storm Karl (2016) to demonstrate the applicability of our clustering procedure. All identified clusters prove to be physically meaningful and describe distinct flavors of inflow, ascent, outflow, and quasi-horizontal motion in Karl’s vicinity. Importantly, the clusters exhibit gradual temporal evolution, which is most notable because the clustering procedure itself does not impose temporal consistency on the clusters. Finally, TC problems are discussed for which the application of the clustering procedures seems to be most fruitful.

The rapidly growing computational demands of deep neural networks require novel hardware designs. Recently, tuneable nanoelectronic devices were developed based on hopping electrons through a network of dopant atoms in silicon. These ‘dopant network processing units’ (DNPUs) are highly energy-efficient and have potentially very high throughput. By adapting the control voltages applied to its electrodes, a single DNPU can solve a variety of linearly non-separable classification problems. However, using a single device has limitations due to the implicit single-node architecture. This paper presents a promising novel approach to neural information processing by introducing DNPUs as high-capacity neurons and moving from a single to a multi-neuron framework. By implementing and testing a small multi-DNPU classifier in hardware, we show that feed-forward DNPU networks improve the performance of a single DNPU from 77% to 94% test accuracy on a binary classification task with concentric classes on a plane. Furthermore, motivated by the integration of DNPUs with memristor crossbar arrays, we study the potential of using DNPUs in combination with linear layers. We show by simulation that an MNIST classifier with only 10 DNPU nodes achieves over 96% test accuracy. Our results pave the road towards hardware neural network emulators that offer atomic-scale information processing with low latency and energy consumption.

Extratropical transition (ET) of tropical cyclones involves distinct changes of the cyclone’s structure that are not yet well understood. This study presents for the first time a comprehensive Lagrangian description of structure change near the inner core. A large sample of trajectories is computed from a convection-permitting numerical simulation of the ET of Tropical Storm Karl (2016). Three main airstreams are considered: those associated with the inner-core convection, inner-core descent, and the developing warm conveyor belt. Analysis of these airstreams is performed both in thermodynamic and physical space. Prior to ET, Karl is embedded in weak vertical wind shear and its intensity is impeded by excessive detrainment from the inner-core convection. At the start of ET, vertical shear increases and Karl intensifies, which is attributable to reduced detrainment and thus to the formation of a well-defined outflow layer. During ET, the thermodynamic changes of the environment impact Karl’s inner-core convection predominantly by a decrease of θe values in the inflow layer. Notably, notwithstanding Karl’s weak intensity, its inner core acts as a “containment vessel” that transports high-θe air into the increasingly hostile environment. Inner-core descent has two origins: (i) mostly from upshear-left above 4-km height in the environment and (ii) boundary layer air that ascends in the inner core first and then descends, performing rollercoaster-like trajectories. At the end of the tropical phase of ET, the developing warm conveyor belt comprises air masses from several different source regions, and only partly from the cyclone’s developing warm sector, as expected for extratropical cyclones.

Plastics are indispensable to modern life, but their widespread use has created an environmental crisis due to inefficient waste management. Mixed plastic waste, comprising diverse polymers, presents significant recycling challenges due to the high costs of sorting and processing, leading to ecosystem accumulation and harmful by-product generation. This study addresses this issue by engineering a synthetic bacterial consortium (SBC) designed to degrade mixed plastic monomers. The consortium pairs Escherichia coli Nissle 1917, which uses ethylene glycol (EG), a monomer derived from polyethylene terephthalate (PET), as a carbon source, with Pseudomonas putida KT2440, which metabolizes hexamethylenediamine (HD), a monomer from nylon-6,6, as a nitrogen source. Adaptive evolution of the SBC revealed a novel metabolic interaction where P. putida developed the ability to degrade both EG and HD, while E. coli played a critical role in degrading glycolate, mitigating its by-product toxicity. The evolved cross-feeding pattern enhanced biomass production, metabolic efficiency, and community stability compared to monocultures. The consortium’s performance was validated through flux balance analysis (FBA), high-performance liquid chromatography (HPLC), and growth assays. These findings highlight the potential of cross-feeding SBCs in addressing complex plastic waste, offering a promising avenue for sustainable bioremediation and advancing future polymer degradation strategies.