We report a k-point extension of the second-order co-iterative augmented Hessian (CIAH) algorithm, termed k-CIAH, for Pipek-Mezey (PM) localization of Wannier functions (WFs). By exploiting an efficient evaluation of the Hessian-vector product, k-CIAH achieves O(Nk2n3) scaling in both CPU time and memory, matching that of previously reported first-order k-space approaches while improving upon the O(Nk3n3) scaling of Γ-point CIAH, where Nk denotes the number of k-points sampling the first Brillouin zone and n characterizes the unit-cell size. Benchmark calculations on a diverse set of solids─including insulators, semiconductors, metals, and surfaces─demonstrate the fast and robust convergence of k-CIAH-based PMWF optimization, which yields an overall computational efficiency approximately 2-3-fold higher than first-order k-space methods and orders of magnitude higher than Γ-point CIAH for localizing 1000-5000 orbitals. The quality of the resulting PMWFs is further validated by accurate electronic band structures obtained via PMWF-based Wannier interpolation.

Abundant new nanomaterials are developed every year to enhance traditional radiotherapy by improving therapeutic response and reducing side effects. However, current pre-clinical models for evaluating these nanomaterials suffer from low fidelity (e.g., 2D cellular assays) or high cost (e.g., model animals). Considering that tumor spheroids can proliferate on a large scale and at low cost while possessing complex architectures that replicate in vivo solid tumors, in this work, we developed a microfluidic platform integrated with uniform tumor spheroids for the fast and reliable evaluation of radio-therapeutic nanomaterials. Utilizing a cascade of two microfluidic chips (G-chip and T-chip), our platform enabled fast generation, long-term cultivation, and precise pre-treatment of uniform tumor spheroids. The narrow size distribution and high viability of the tumor spheroids originating from the droplet microfluidic G-chip ensured the subsequent evaluation was reproducible and reliable. By regulating the concentration gradient formed in the T-chip through controlling fluidic flow rates, tumor spheroids can be pretreated with different concentrations of radio-therapeutic nanomaterials precisely and simultaneously. Herein, a previously reported radio-therapeutic nanoenzyme was employed to implement radiotherapy evaluation based on the platform as a proof of concept, producing results with high fidelity to in vivo solid tumor. This microfluidic platform shows great potential for the straightforward, reliable, and cost-effective evaluation of radiotherapy nanomaterials.

Statics and kinematics are often viewed as intertwined branches of mechanics. The principle of virtual work indicates this interconnection: For a system in static equilibrium, the total work performed by all forces during any virtual displacement must equal zero. In this work, we make the interplay between statics and kinematics more explicit by focusing on two engineering structures-tensegrity and origami. Specifically, we demonstrate a quantitative duality relationship between the states of self-stress of tensegrity and the infinitesimal mechanisms of origami. More importantly, we show that this duality remains invariant under nondegenerate linear transformations applied to the tensegrity and origami configurations. Furthermore, we establish that the stability property of tensegrity, particularly superstability, is preserved under such transformations. We apply the invariant duality theory to tensegrity and origami structures with prismatic and polyhedral geometries, illustrating its broad applicability. Such duality is also applicable to the fast generation of irregular, three-dimensional architected materials and structures.

Wheat (Triticum aestivum L.) is one of the most cultivated crops in the world, but production is often affected by drought. The wheat chromosome 4A contains several quantitative trait loci (QTL) associated with drought tolerance and yield-related traits, making it a valuable target for genetic improvement. In this study, we developed near-isogenic lines (NILs) carrying qDT.4A.1, a major QTL for yield using a fast generation cycling system (FGCS) and characterized these NILs for grain yield and thousand-grain weight (TGW) under drought stress and control conditions. We identified a single nucleotide polymorphism (SNP) marker Kukri_c27037_112, which showed a consistent genotype-phenotype associations across two NIL pairs. This marker is linked to four candidate genes encoding a RING-finger E3 ubiquitin ligase, a receptor kinase, and a protein kinase family protein involved in drought stress response and pathways. In silico expression analyses revealed upregulation of these genes in grain tissue under drought conditions, supporting their potential role in grain development and yield formation during drought stress conditions. The identified SNP marker and its associated candidate genes are potential resources in marker-assisted selection and fine mapping pending further validation and functional studies. Our results provide valuable genomic resources, laying the foundation for the development of drought-tolerant wheat varieties and highlighting chromosome 4A as a key region governing drought tolerance.

Medical image synthesis is an important topic for both clinical and research applications. Recently, diffusion models have become a leading approach in this area. Despite their strengths, many existing methods struggle with (1) limited generalizability, only working for specific body regions or voxel spacings, (2) slow inference, which is a common issue for diffusion models, and (3) weak alignment with input conditions, which is a critical issue for medical imaging. MAISI, a previously proposed framework, addresses generalizability issues but still suffers from slow inference and limited condition consistency. In this work, we present MAISI-v2, the first accelerated 3D medical image synthesis framework that integrates rectified flow to enable fast and high-quality generation. To further enhance condition fidelity, we introduce a novel region-specific contrastive loss to improve sensitivity to the region of interest. Our experiments show that MAISI-v2 can achieve state-of-the-art image quality with 33× acceleration for latent diffusion models. We also conducted a downstream segmentation experiment to show that the synthetic images can be used for data augmentation. We release our code, training details, model weights, and a GUI demo to facilitate reproducibility and promote further development within the community.

Symmetry-breaking charge separation (SB-CS) shows enormous potential applications in solar energy conversion. Nonetheless, the realization of fast CS coupled with slow charge recombination (CR), which is crucial to achieving its application, remains a great challenge. To address this challenge, we have synthesized an anthracene dimer and a trimer by increasing the number of structural units to achieve fast generation and slow recombination of the SB-CS state. Transient absorption spectra show that these two oligomers could undergo the SB-CS process, even in low-polarity solvents. In the same solvent, the SB-CS rate of the trimer is 1.5-fold faster than that of the dimer, while its recombination rate is slowed down (from 1/17 ns-1 to 1/20 ns-1). The rate ratio between SB-CS formation and recombination in biph-trimer reaches an impressive ∼4000 in DMF, which is the highest record observed in the anthracene derivative system. These results suggest that increasing the structural unit number in oligomers may be an effective method to achieve fast SB-CS and slow CR.

Cyclic peptides are an emerging therapeutic modality, with recent computational efforts focusing on the design of cyclic peptides that predominantly adopt a single conformation. However, many cyclic peptides adopt multiple conformations in solution, existing as structural ensembles. This conformational flexibility is often integral to their function: chameleonic switching between alternative states can enhance membrane permeability, and specific conformations may be required for molecular recognition and binding. Consequently, the ability to predict their structural ensembles is crucial for advancing the de novo design of cyclic peptide therapeutics. Here, we introduce diffusion models to efficiently and accurately predict structural ensembles of mixed-chirality cyclic peptides. The models are trained directly on molecular dynamics (MD) simulation data; in particular, each frame of the simulation becomes a single training instance in which a structure is represented as sine and cosine values of backbone dihedral angles. The trained diffusion model can not only generate MD-quality structures of cyclic peptides, but also the generated structures follow the Boltzmann distribution sampled in the MD simulation, enabling a deeper understanding of the physicochemical basis of cyclic peptide properties and allowing efficient computational design of cyclic peptides targeting biologically relevant systems.

Quantitative Structure-Activity Relationship (QSAR) is a computer-based tool that depicts empirical aspects in drug modeling. While it was limited to physical organic chemistry for the past fifty years, QSAR modeling has been diversified and has become more challenging, especially in drug design. From physicochemical property prediction to toxicity predictions, ADME properties, and data mining, it has changed the perspective in drug designing. This innovation was much needed in drug design due to the increasing complexity of the process, which demands more proficient tools and a lower probability of errors. However, when it comes to challenges like predicting potency, fast structure-activity generation, and series design, QSAR has much to offer in the near future. This article aims to give an overview of modern drug chemistry and the importance of various QSAR approaches in drug designing across various fields. The present manuscript discusses the application of QSAR methods in drug design and development, along with a historical overview of various QSAR approaches, supported by relevant examples.

A fast gray-box adversarial example generation algorithm based on FakeBob

The field of neural rendering has seen remarkable progress, driven by advancements in generative models and differentiable rendering techniques. While 2D diffusion has achieved notable success, the development of a unified 3D diffusion pipeline remains an open challenge. This paper presents a novel framework, LN3Diff++, designed to bridge this gap and facilitate fast, high-quality, and versatile conditional 3D generation. Our method leverages a 3D-aware architecture and a variational autoencoder (VAE) to encode input image(s) into a structured, compact 3D latent space. The latent representation is then decoded by a transformer-based decoder into a high-capacity 3D neural field. By training a diffusion model on this 3D-aware latent space, our method achieves superior performance for category-specific 3D generation on ShapeNet and FFHQ, as well as category-free image/text-conditioned 3D generation over Objaverse. Moreover, it surpasses existing 3D diffusion methods in inference speed, requiring no per-instance optimization.

We propose a hybrid quantum device based on the graphene Josephson junctions, where the vibrational degrees of freedom of a graphene membrane couple to the superconducting circuits. The flexural mode-controlled tunneling of the Cooper pairs introduces a strong and tunable coupling even at the zero-point fluctuation level. By employing this interaction, we show that a parametric process can be efficiently implemented. We then investigate foundational and technological applications of our hybrid device empowered by nonlinear interactions, with fast generation of nonclassical mechanical states, and critically enhanced quantum sensing under suitable quantum control. Our work provides the possibility of employing the graphene motional degree of freedom for quantum information processing in circuit quantum nanomechanical structures.

Organic radicals are highly important, but it is still challenging to produce air-stable and photo-induced radicals. Herein, a series of triarylamine derivatives (compounds 1-6) is designed to tune their molecular stacking and then their capability in photo-induced electron transfer and radical stabilization. It is noted that, in contrast to the other compounds, a robust hydrogen-bonded crystal network is observed in compound 1 with parallel alignments in two dimensions. This network is beneficial to alter its electronic and spatial structures, leading to enhanced inter-molecular electronic transfer. This can finally facilitate the rapid photo-response and high radical stability. As observed, upon irradiation, compound 1 exhibits a fast radical generation in solution (5-min to saturation), showing a 22.5-folds enhancement in near-infrared absorption and 8-folds increase in emission. Remarkably, these radical signals can also be obtained in crystals and detected in the solid state even after one month. Interestingly, it also exhibits efficient fluorescence emission in the NIR region, also high capability to generate reactive oxygen species and heat under irradiation. Therefore, as a proof of concept, these radical-based nanoparticles are prepared via nanoprecipitation for fluorescence imaging and photo-therapy simultaneously both in vitro and in vivo, highlighting their great potential in phototheranostics.

ABSTRACT The aim of this study was to develop, for the first time, an analytical methodology for the simultaneous determination of inorganic arsenic and antimony in saline water samples by hydride generation atomic absorption spectrometry (HG-AAS), employing a fast sequential approach. To enable the joint determination of both analytes in a single run, a compromise condition was established to ensure their quantitative reduction to the trivalent state. This reduction was achieved using thiourea as a reducing agent at a concentration of 1.0% (m/v) in a medium containing 5 mol L−1 HCl. Once the pre-reduction step was optimised, the measurement conditions of the system were adjusted to allow the efficient formation of volatile hydrides of As and Sb. Optimal responses were obtained using a 0.5% (m/v) NaBH4 solution prepared in 0.5% (m/v) NaOH and a 5 mol L−1 HCl reagent solution to maintain the required acidity. Under the optimised conditions, the limits of quantification for inorganic As and Sb were 1.5 and 1.1 µg L−1, respectively, values considerably lower than the maximum limits recommended by national and international regulations. Recovery tests performed with samples spiked at 5.0 and 10 µg L−1 yielded recoveries ranging from 95 to 120% for total inorganic As and from 82 to 120% for total inorganic Sb, confirming the accuracy and applicability of the proposed methodology.

A fast and compact pulse power generator has been developed for creating conditions for high-pressure dynamic compression in solids as well as in pulsed plasma. The generator consists of four identical capacitor-switch units connected concentrically around a coaxial load section. Each capacitor-switch unit consists of a low-inductance configuration of a 2100 nF/55kV capacitor with coaxial terminals and an ambient air-operated, planar, 5-gap, 4-channel multi-gap, multi-channel spark gap switch to exhibit unit impedance up to 0.25 Ω. For a maximum charging voltage of 50kV, the energy stored in the generator is ∼10 kJ. A negative-polarity HV pulse of magnitude -68 kV and fall time ∼56ns is used to trigger all four switches simultaneously into the load section. Under short-circuit characterization at 50kV, the developed generator has been found to generate current pulses of magnitude 735kA with a rise-time (10%-90%) of 600ns. The overall dimensions of the generator are 985 × 985 × 760mm3, and the total weight is ∼500kg, thereby making it a simple, low-cost apparatus for conducting dynamic compression experiments under laboratory conditions. Ambient air operation of the generator up to 50kV and its extremely low generator impedance of ∼0.076 Ω make it a robust and small-footprint pulse power driver suited for experiments on low-impedance loads in a laboratory environment.

Abstract This paper proposes a high-performance, low-power Multiply-Accumulate (MAC) architecture leveraging a Dynamic Reconfiguration with Parallel Partial Evaluation (DRPPE) framework integrated with a 16×16 Vedic multiplier and a flag-driven Selective Partial Sum (SPS) adder. The design addresses the critical need for speed, area efficiency, and dynamic power reduction in digital signal processing (DSP), artificial intelligence (AI), and FPGA-based systems. By incorporating Urdhva-Tiryakbhyam Sutra-based Vedic multiplication, the architecture enables fast partial product generation and low-depth logic realization. The novel flag control mechanism intelligently suppresses redundant carry propagation during accumulation, thereby minimizing switching activity and improving energy efficiency. The proposed design was implemented in Verilog HDL and synthesized on a Virtex-5 FPGA using Xilinx ISE 14.7. Simulation results demonstrate a critical path delay of 2.05 ns, maximum operating frequency of 465.1 MHz, and total power consumption of 52.5 mW, significantly outperforming conventional MAC units in speed and resource utilization. Comparative analysis with recent literature confirms the design's effectiveness in delivering accuracy, configurability, and high-throughput performance suitable for next-generation embedded computing platforms.

Coronary artery segmentation in non-contrast calcium scoring CT images using deep learning.

Enhancement of DNA recovery in rapid DNA technology: A novel pre-extraction protocol.

Abstract Large-scale Greenberger–Horne–Zeilinger (GHZ) state is useful for quantum technologies but difficult to be prepared. Here, we propose fast measurement-based preparation of large-scale GHZ states by a four-qubit quantum phase gate with nuclear-spin qubits of alkaline-earth-like atoms, which is named as quantum ferromagnetic gate due to its analogy to the alignment of molecular magnetic moments in a classical magnet. A high-fidelity Rydberg-mediated QFG can be realized in a time of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mrow> <mml:mn>6</mml:mn> <mml:mi>π</mml:mi> <mml:mrow> <mml:mo>/</mml:mo> </mml:mrow> <mml:msub> <mml:mi mathvariant="normal">Ω</mml:mi> <mml:mrow> <mml:mtext>m</mml:mtext> </mml:mrow> </mml:msub> </mml:mrow> </mml:math> with <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mrow> <mml:msub> <mml:mi mathvariant="normal">Ω</mml:mi> <mml:mrow> <mml:mtext>m</mml:mtext> </mml:mrow> </mml:msub> </mml:mrow> </mml:math> the maximal Rydberg Rabi frequency. From a product state of three data atoms and one ancilla atom, a gluing circuit with one QFG, two single-qubit gates, and a projective measurement of the ancilla can generate a 3-qubit GHZ state, and repetition of this gluing circuit can lead to 9, 27, 81, 243 <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mrow> <mml:mo>,</mml:mo> <mml:mo>…</mml:mo> </mml:mrow> </mml:math> -qubit GHZ states. Analyses based on currently available techniques show that a 243-qubit GHZ state is realizable, and more qubits can be entangled with higher detection fidelity.

Photoacoustic computed tomography (PACT) is a promising imaging modality that combines the advantages of optical contrast with ultrasound detection. Utilizing ultrasound transducers with larger surface areas can improve detection sensitivity. However, when computationally efficient analytic reconstruction methods that neglect the spatial impulse responses (SIRs) of the transducer are employed, the spatial resolution of the reconstructed images will be compromised. Although optimization-based reconstruction methods can explicitly account for SIR effects, their computational cost is generally high, particularly in three-dimensional (3D) applications. To address the need for accurate but rapid 3D PACT image reconstruction, this study presents a framework for establishing a learned SIR compensation method that operates in the data domain. The learned compensation method maps SIR-corrupted PACT measurement data to compensated data that would have been recorded by idealized point-like transducers. Subsequently, the compensated data can be used with a computationally efficient reconstruction method that neglects SIR effects. Two variants of the learned compensation model are investigated that employ a U-Net model and a specifically designed, physics-inspired model, referred to as Deconv-Net. A fast and analytical training data generation procedure is also a component of the presented framework. The framework is rigorously validated in virtual imaging studies, demonstrating resolution improvement and robustness to noise variations, object complexity, and sound speed heterogeneity. When applied to in-vivo breast imaging data, the learned compensation models revealed fine structures that had been obscured by SIR-induced artifacts. To our knowledge, this is the first demonstration of learned SIR compensation in 3D PACT imaging.

We demonstrate a method to generate application-ready truly random bits from a magnetic tunnel junction driven by a Field-Programmable Gate Array (FPGA). We implement a real-time feedback loop that stabilizes the switching probability near 50% and apply an XOR operation, both on the FPGA, to suppress short-term correlations, together mitigating long-term drift and bias in the bitstream. This combined approach passes the NIST Statistical Test Suite for random bit generation at 5 Mb/s without post-processing, providing a practical hardware solution for fast and reliable true random number generation. Beyond cryptographic applications, these capabilities open opportunities for stochastic hardware accelerators, probabilistic computing, and large-scale modeling where real-time access to unbiased randomness is essential.