We have developed a hypersonic high-order, high-performance code (H^3PC) utilizing the “Trixi.jl" framework in order to simulate both non-reactive and chemically reactive compressible Euler and Navier-Stokes equations for complex three-dimensional geometries. H^3PC is parallel on CPU platforms and can perform exascale parallel computations of hypersonic turbulent flows. The numerical approach is based on the discontinuous Galerkin spectral element method, satisfying the entropy and energy stability conditions for the Euler equations. H^3PC can perform simulations of high-speed flows from subsonic to hypersonic speeds based on frozen, equilibrium, and non-equilibrium chemistry modeling of the gas mixture, using the , which is a Julia package developed to wrap the C++-based Mutation++ library. H^3PC can also perform parallel adaptive mesh refinement for two- and three-dimensional Euler and Navier-Stokes discretizations with non-conforming elements. In this study, we first demonstrate the successful integration of Mutation++ into the H^3PC solver, and then verify its accuracy through simulations of Taylor-Green vortex flow, supersonic flow past a square and circular cylinder, and hypersonic P8-inlet.
Purpose Patients with end-stage renal disease (ESRD) have high rates of pathologies, which may necessitate finger amputation, such as infection or vascular insufficiency. Still, little work has examined its significance in upper-extremity amputation. This study aimed to explore ESRD as a predictor of postoperative sequelae in the setting of finger amputation. Methods A retrospective cohort study was performed using the TriNetX database to identify patients with finger amputation through the phalanx, metacarpal, or distal to the carpometacarpal with or without preexisting ESRD or dependence on dialysis. Propensity score matching was used to control for risk factors such as diabetes, atherosclerosis, peripheral vascular disease hypertension, among others. Outcomes included return to operating room or death within 1-year, subsequent ray amputation, proximal upper-extremity amputation, and other common sequelae of finger amputation. Results After propensity score matching, 1,619 patients without and 1,619 patients with ESRD or dialysis dependence before index finger amputation were in each cohort. More patients with ESRD returned to the operating room within 1 year for debridement or secondary closure of dehisced wound (334 [20.63%] vs 217 [13.40%]). Furthermore, 338 (20.88%) non-ESRD amputees and 536 (33.11%) ESRD amputees underwent repeat or subsequent ray amputation. Similarly, patients with ESRD demonstrated higher rates of proximal upper extremity amputation (ie, at the hand, wrist, or forearm) following index surgery: 296 (18.28%) patients without ESRD versus 481 (29.71%) patients with ESRD. ESRD amputees also developed significantly higher rates of chronic osteomyelitis (290 [17.91%] vs 219 [13.53%], P < .001) as well as death within 1 year (375 [23.16%] vs 121 [7.47%], P < .001). Conclusions Patients with ESRD experience higher rates of subsequent proximal upper extremity amputation, subsequent ray amputation, chronic osteomyelitis, return to the operating room, and death. Further work expositing the pathophysiological mechanism of these associations may assist hand surgeons in preoperative optimization of these patients. Type of study/level of evidence Prognostic IIb.
Deep generative models are increasingly powerful tools for the in silico design of novel proteins. Recently, a family of generative models called diffusion models has demonstrated the ability to generate biologically plausible proteins that are dissimilar to any actual proteins seen in nature, enabling unprecedented capability and control in de novo protein design. However, current state-of-the-art diffusion models generate protein structures, which limits the scope of their training data and restricts generations to a small and biased subset of protein design space. Here, we introduce a general-purpose diffusion framework, EvoDiff, that combines evolutionary-scale data with the distinct conditioning capabilities of diffusion models for controllable protein generation in sequence space. EvoDiff generates high-fidelity, diverse, and structurally-plausible proteins that cover natural sequence and functional space. We show experimentally that EvoDiff generations express, fold, and exhibit expected secondary structure elements. Critically, EvoDiff can generate proteins inaccessible to structure-based models, such as those with disordered regions, while maintaining the ability to design scaffolds for functional structural motifs. We validate the universality of our sequence-based formulation by experimentally characterizing intrinsically-disordered mitochondrial targeting signals, metal-binding proteins, and protein binders designed using EvoDiff. We envision that EvoDiff will expand capabilities in protein engineering beyond the structure-function paradigm toward programmable, sequence-first design.
Joint Embedding Predictive Architectures (JEPAs) offer a compelling framework for learning world models in compact latent spaces, yet existing methods remain fragile, relying on complex multi-term losses, exponential moving averages, pre-trained encoders, or auxiliary supervision to avoid representation collapse. In this work, we introduce LeWorldModel (LeWM), the first JEPA that trains stably end-to-end from raw pixels using only two loss terms: a next-embedding prediction loss and a regularizer enforcing Gaussian-distributed latent embeddings. This reduces tunable loss hyperparameters from six to one compared to the only existing end-to-end alternative. With 15M parameters trainable on a single GPU in a few hours, LeWM plans up to 48x faster than foundation-model-based world models while remaining competitive across diverse 2D and 3D control tasks. Beyond control, we show that LeWM's latent space encodes meaningful physical structure through probing of physical quantities. Surprise evaluation confirms that the model reliably detects physically implausible events.
NASA's Curiosity rover is exploring a 5 km tall sedimentary mound that is hypothesized to record the transition from a warm and wet (phyllosilicate-rich) to a cold and drier (sulfate-rich) Mars. Evidence of magnesium sulfate-bearing rock has shown that Curiosity has crossed through this phyllosilicate-sulfate transition. Recently, Curiosity arrived at the Amapari Marker Band, a darker, indurated unit that can be traced laterally for tens of kilometers in orbiter images. Here, Curiosity found evidence for a very broad lake, and bedforms interpreted as wave-ripple laminated sedimentary rock that likely was deposited in shallow water in the explored location, before becoming a deeper lake. These rocks are enriched in Fe, Mn, and Zn which has major implications for groundwater paleohydrology in Gale crater. Three formation hypotheses are considered: concretion formation during early diagenetic alteration of shallow lake sediments, laterization or leaching of the sediments, and addition of Fe, Mn, and Zn by a mildly acidic and reducing groundwater interacting with a redox and/or pH front in a stratified lake. The preferred interpretation of the metal enrichments within the Amapari Marker band sedimentary rocks is that they formed in a shallow water environment at a redox and/or pH front within the ripple unit, which drove precipitation and concentration of metals. If the enrichments are due to groundwater alteration, these processes could link subsurface and surface environments. Water and the presence of high amounts of redox sensitive elements and other metals are favorable indicators for habitability.
