Rapid reaction-thermal diffusion during frontal polymerization (FP) with variations in initial and boundary conditions destabilizes the planar mode of front propagation, leading to spatially varying complex hierarchical patterns in polymeric materials. Although modern reaction-diffusion models can predict the patterns resulting from unstable FP, the inverse design of patterns, which aims to retrieve process conditions that produce a desired pattern, remains an open challenge due to the nonunique and nonintuitive mapping between process conditions and patterns. In this work, we propose a novel probabilistic generative model named univariate conditional variational autoencoder (UcVAE) for the inverse design of hierarchical patterns in FP-based manufacturing. Unlike the cVAE, which encodes both the design space and the design target, the UcVAE encodes only the design space. In the encoder of the UcVAE, the number of training parameters is significantly reduced compared to the cVAE, resulting in a shorter training time while maintaining comparable performance. Given desired pattern images, the trained UcVAE can generate multiple process condition solutions that produce high-fidelity hierarchical patterns.

This paper investigates the dynamics of non-autonomous cooperative systems of difference equations with asymptotically constant coefficients. We are mainly interested in global attractivity results for such systems and the application of such results to evolutionary population cooperation models. We use two methods to extend the global attractivity results for autonomous cooperative systems to related non-autonomous cooperative systems which appear in recent problems in evolutionary dynamics.

Plant pathogenic fungi have evolved different strategies to interfere with plant defense mechanisms. The well described fungal plant pathogen Fusarium graminearum is not only able to produce trichothecene toxins like deoxynivalenol, but also the plant hormone auxin. Highly elevated levels of auxin and auxin derivatives such as IAA-glucoside or IAA amino-acid conjugates were observed in wheat cultivar Apogee infected with F. graminearum. We report that F. graminearum is able to cleave tryptamine-derived hydroxycinnamic acid amides, e.g. the defense compound coumaroyl-tryptamine. In this study we investigated copper amine-oxidases, candidate genes for auxin biosynthesis converting tryptamine into the IAA precursor indole-3-acetyldehyde. After consecutive knock outs of all seven copper amine oxidases the resulting septuple knock out strain had strongly reduced ability to produce auxin. Virulence of the septuple mutant was significantly impaired while DON production in planta was comparable to the wild type. We conclude that F. graminearum, often presumed to be a simple nectrotroph, has a biotrophic phase and is able to employ plant defense compounds by converting them into defense suppressing auxin.

Uloga ustavnog sudstva u svakoj savremenoj demokratski uređenoj državi od nemjerljivog je značaja za očuvanje njenog ustavnopravnog poretka i osnovnih principa pravne države i vladavine prava, ustavnosti i zakonitosti. Pozicija Ustavnog suda Bosne i Hercegovine u sistemu državnih institucija zauzima posebno mjesto, zbog specifičnosti njenog državnog uređenja, stepena demokratskog razvoja i opredijeljenosti političkih aktera ka poštivanju uvodno navedenih principa. Da bi Ustavni sud Bosne i Hercegovine odgovorio svojoj zahtjevnoj ulozi “čuvara” Ustava Bosne i Hercegovine i njenog ustavnopravnog poretka, posebno je važno osigurati da u sastav Ustavnog suda Bosne i Hercegovine budu izabrani samo i isključivo nezavisni, nepristrasni, stručni i profesionalni pravnici visokih moralnih kvaliteta, bez obzira na njihovu etničku pripadnost, oslobođeni mogućnosti bilo kakvih političkih ili drugih uticaja na njihove stavove, mišljenja i odluke, osigurati objektivne i nezavisne procedure donošenja odluka bez bilo kakvih institucionalnih ili vaninstitucionalnih uticaja i pritisaka, posebno kada su u pitanju procedure donošenja odluka koje se ne mogu i ne smiju vezati za etničku ili entitetsku pripadnost pojedinog sudije Ustavnog suda Bosne i Hercegovine. Upravo kroz mehanizme etnicizacije i politizacije pitanja sastava, izbora sudija i načina donošenja odluka Ustavnog suda Bosne i Hercegovine, a posmatrajući ulogu Ustavnog suda Bosne i Hercegovine u očuvanju državnosti i državnopravnog kontinuiteta, svjedoci smo očiglednom podrivanju ustavnopravnog poretka države Bosne i Hercegovine.

Heat dissipation and thermal management is a rising concern for nanoelectronic devices and threatens to curtail their adoption in integrated circuits, sensors, and energy converters. Joule heating due to dissipation in the channel region of nanoelectronic devices causes increased temperature and may lead to mobility degradation and long-term reliability issues. Here we study thermal transport and cross-plane thermal boundary conductance in a variety of “beyond graphene” 2D materials and few-layer stacks on several amorphous and crystalline substrates using a combination of first principles methods and Boltzmann transport of phonons. We employ machine learning to accelerate the discovery of 2D-substrate pairings with enhanced thermal conductance. Beyond that, we couple electronic and thermal transport to study dissipation in field effect MOS transistors and show that heat dissipation is non-uniform and that self-heating reduces mobility. We find that judicious selection of the number of layers and substrate can significantly reduce the deleterious effects of Joule heating.

Raman thermometry has gained immense popularity for probing the thermal properties of nanostructured materials due to its excellent spatial resolution and lack of contact error; however, it has a key weakness in its temperature resolution. In this work, we aim to improve the temperature resolution of Raman thermometry through training neural networks to track the locations, widths, and relative heights of multiple peaks at once. We find that in training a multilayer perceptron on 13 pixel values representing the Raman peak of silicon, the variance and standard deviation in thermal conductivity predictions can be reduced as compared to those resulting from the predominant method of tracking the peak location as it shifts with temperature. We expect that this work may contribute to greater accuracy of thermal measurements from non-contact Raman-based techniques and thereby improve the consensus on the thermal properties of 2D materials.

Despite their potential for miniaturization, electronic devices made of 2D materials face thermal management challenges due to their reduced dimensionality, which can limit their efficiency and lifespan. Low thermal boundary conductance (TBC) is one major limiting factor in realizing efficient heat transfer to the substrate. Due to the roughness at the interface, the adhesion of 2D materials to their substrates tend to be weak, resulting in low TBC. Therefore, to improve heat flow from the 2D material, we need to discover novel ways of increasing TBC. In this study, we have used a numerical model combined with first-principles DFPT simulations to investigate a possible method to increase TBC using an electrostatic field due to gate voltage. Our study shows that electrostatic pressure can be used to effectively enhance TBC for an interface formed by a 2D material and a rough substrate. We find that electrostatic pressure can improve TBC by more than 300 % when an electric field of 3 V/nm is applied. This is due to an improvement in the vdW spring coupling constant, which shows a more than two-fold increase when a substrate roughness of 1.6 nm and correlation length of 10.8 nm, 2D-material's bending stiffness of 1.5 eV, and adhesion energy of 0.1 $J/m^{2}$ were used. We show that TBC is enhanced more when the substrate has a large roughness and small correlation length, and the $2D$ material has a large bending stiffness. This is because a stiff 2D sheet resist bending when voltage/pressure is applied, thus causing it to press more on the roughness peaks, resulting in a tremendous increase in the coupling constants at the peaks in the atomically rough surface of the substrate. However, a flexible 2D material can easily bend to conform to the topography of the rough substrate when voltage/pressure is applied, which makes the coupling constants across the interface more uniform. Here we show that TBC is enhanced more when adhesion is weak because a weak vdW bond is easily compressed by external pressure. Therefore, our study provides valuable information that can be applied in designing electronic devices with efficient heat management by using gate voltage, substrate roughness combined with the mechanical properties.

Conjugated polymers (CP) are frequently doped to modulate their transport and optical properties. Doping alters the intrinsic Gaussian density of states (DOS) by adding Coulomb energy and inducing an exponential tail. Changes in transport or optical properties are mainly tracked back to changes in DOS and carrier hopping rates. Conductivity shows a power-law like increase and the Seebeck coefficient a decrease with carrier concentration. This results in a trade-off between transport properties with doping. However, their modification with doping is still not well understood. Here we show that capture transport and optical properties of doped CPs, by developing a tight-binding Hamiltonian that includes dopant-induced energetic disorder (DID) via Coulomb interactions. We utilize perturbation theory to calculate transition rates between wavefunctions from the calculated eigenenergies and eigenfunctions. With the obtained transition rates, we solve Pauli master equation for occupational probabilities to compute transport properties of doped CPs. Additionally, we capture optical absorption features by simply simulating the joint DOS and IR absorption features via simulated AC conductivity. We anticipate our work to significantly contribute to understanding of underlying transport and optical physics of doped CPs.

Inflammatory bowel disease (IBD), encompassing Crohn’s disease and ulcerative colitis, is a chronic and relapsing condition characterized by persistent inflammation of the gastrointestinal tract. The complex pathogenesis of IBD involves a combination of genetic, environmental, and immune factors, which complicates the achievement of long-term remission. Lower abdominal pain, stomach cramps, blood in stool, chronic diarrhea, fatigue, and unexpected weight loss are common presenting symptoms. Despite the range of therapies and medications, including anti-inflammatory and anti-diarrheal drugs, immunosuppressants, antibiotics, and analgesics aimed at managing symptoms and controlling inflammation, a definitive cure for IBD remains elusive. Current therapy targets inflammation, mainly cytokines, inflammatory receptors, and immune cells, however, there is a need for novel targets to improve clinical outcomes. To identify novel targets and interactions among various factors, we performed a network analysis using various cytokines, TLRs, and NLRP3 inflammasome as inputs. This analysis revealed orosomucoid-like protein 3/ORMDL sphingolipid biosynthesis regulator 3 (ORMDL3) as a central hub gene interacting with multiple factors. While the role of ORMDL3 in IBD pathogenesis is not well-established, our findings and existing literature suggest that ORMDL3 plays a role in inflammation, impaired mitochondrial function, and disrupted autophagy, all contributing to the disease progression. Given its central role in these pathogenic processes, targeting ORMDL3 presents a promising therapeutic target. Modulating ORMDL3 activity could alleviate inflammation, restore mitochondrial function, and enhance autophagy, potentially leading to more effective treatments and improved outcomes for IBD patients.

The high computational costs of large language models (LLMs) have led to a flurry of research on LLM compression, via methods such as quantization, sparsification, or structured pruning. A new frontier in this area is given by \emph{dynamic, non-uniform} compression methods, which adjust the compression levels (e.g., sparsity) per-block or even per-layer in order to minimize accuracy loss, while guaranteeing a global compression threshold. Yet, current methods rely on heuristics for identifying the"importance"of a given layer towards the loss, based on assumptions such as \emph{error monotonicity}, i.e. that the end-to-end model compression error is proportional to the sum of layer-wise errors. In this paper, we revisit this area, and propose a new and general approach for dynamic compression that is provably optimal in a given input range. We begin from the motivating observation that, in general, \emph{error monotonicity does not hold for LLMs}: compressed models with lower sum of per-layer errors can perform \emph{worse} than models with higher error sums. To address this, we propose a new general evolutionary framework for dynamic LLM compression called EvoPress, which has provable convergence, and low sample and evaluation complexity. We show that these theoretical guarantees lead to highly competitive practical performance for dynamic compression of Llama, Mistral and Phi models. Via EvoPress, we set new state-of-the-art results across all compression approaches: structural pruning (block/layer dropping), unstructured sparsity, as well as quantization with dynamic bitwidths. Our code is available at https://github.com/IST-DASLab/EvoPress.