Stochastic diffusion processes are pervasive in nature, from the seemingly erratic Brownian motion to the complex interactions of synaptically-coupled spiking neurons. Recently, drawing inspiration from Langevin dynamics, neuromorphic diffusion models were proposed and have become one of the major breakthroughs in the field of generative artificial intelligence. Unlike discriminative models that have been well developed to tackle classification or regression tasks, diffusion models as well as other generative models such as ChatGPT aim at creating content based upon contexts learned. However, the more complex algorithms of these models result in high computational costs using today's technologies, creating a bottleneck in their efficiency, and impeding further development. Here, we develop a spintronic voltage-controlled magnetoelectric memory hardware for the neuromorphic diffusion process. The in-memory computing capability of our spintronic devices goes beyond current Von Neumann architecture, where memory and computing units are separated. Together with the non-volatility of magnetic memory, we can achieve high-speed and low-cost computing, which is desirable for the increasing scale of generative models in the current era. We experimentally demonstrate that the hardware-based true random diffusion process can be implemented for image generation and achieve comparable image quality to software-based training as measured by the Frechet inception distance (FID) score, achieving ~10^3 better energy-per-bit-per-area over traditional hardware.
Material strategy is an important approach to resist the foreign body response (FBR) that is commonly associated with implants. Only a handful of promising anti-FBR materials have been reported such as zwitterionic materials, modified alginate, and most recently, poly-β-homoserine. In article number 2007226, Runhui Liu and co-workers discuss anti-FBR materials that can prevent implants from recognition by immune cells and have diverse applications in implantable biomaterials and biomedical devices.
A total synthesis of the Annonaceous acetogenins asiminocin and asiminecin is described. The approach is bidirectional starting from the (S,S)-tartrate derived dialdehyde 7 and the (R)-α-OSEM stannane 6. Addition of 6 to 7 in the presence of InCl3 afforded the bis-adduct, anti-diol 8. The derived tosylate 9 was converted to the bis-tetrahydrofuran core unit 10 upon treatment with TBAF. Selective silylation of one of the two equivalent terminal diol groupings led to the OTBS ether alcohol 11. Oxidation to aldehyde 12 and then InCl3-promoted addition of the (S)-allylic stannane 14 gave the anti adduct 15. Removal of the OH group by reduction of the tosylate 16 with LiBEt3H yielded the SEM ether 17. Hydrogenation of the three double bonds of 17 followed by cleavage of the terminal silyl ether and oxidation afforded aldehyde 20. Conversion to the vinylic iodide 21 followed by Pd(0)-catalyzed coupling with the (S)-alkynyl butenolide 24 gave the asiminocin derivative 25. Selective hydrogenation of the enyne moiety with diimide and cleavage of the SEM protecting groups completed the synthesis of asiminocin (27). Asiminecin (41) was prepared starting from aldehyde 12 and the OTBS allylic stannane 28. Addition of the latter to the former in the presence of InCl3 afforded the anti adduct 29 which was protected as the SEM ether 30. Hydrogenation followed by OTBS cleavage with TBAF and selective silylation of the primary alcohol with TBSCl and Et3N−DMAP led to the secondary alcohol 33. Tosylation and hydrogenolysis with LiEt3BH removed the C30 OTs group affording the SEM ether 35. The remaining steps were carried out along the lines described for asiminocin via the vinyl iodide 38 which was coupled with acetylenic butenolide 24 to afford enyne 39. Selective reduction with diimide and SEM cleavage completed the synthesis.
Potent and selective antifungal agents are urgently needed due to the quick increase of serious invasive fungal infections and the limited antifungal drugs available. Microbial metabolites have been a rich source of antimicrobial agents and have inspired the authors to design and obtain potent and selective antifungal agents, poly(DL-diaminopropionic acid) (PDAP) from the ring-opening polymerization of β-amino acid N-thiocarboxyanhydrides, by mimicking ε-poly-lysine. PDAP kills fungal cells by penetrating the fungal cytoplasm, generating reactive oxygen, and inducing fungal apoptosis. The optimal PDAP displays potent antifungal activity with minimum inhibitory concentration as low as 0.4 µg mL
In this study, a novel nanoscale zero-valent iron/chitosan/attapulgite (nZVI/CS/APT) composite was successfully prepared and was employed to remove Naphthol Green B (NGB) from water.Compared to the individual components, the removal performance of the nZVI/CS/APT composite was observed to be dramatically improved.Further, the surface morphology of the composite was characterized, and the possible degradation mechanism was proposed.The results indicated that nanoscale zero-valent iron (nZVI) demonstrated high dispersity and reaction activity after supported by CS/APT.NGB was observed to be degraded into hypotoxic or non-toxic small molecules by nZVI/CS/APT.Further, the critical variables affecting the NGB removal efficiency were screened by the Plackett-Burman design.The interactive effects of the different factors and perfect removal conditions were further elucidated using the response surface methodology.The values of m Fe /m CS-APT , adsorbent dosage, and pH were determined to be the main influencing factors.The average removal efficiency of NGB was determined to be 97.34%(three replicated experiments), which was almost similar to 98.89% predicted by the model.These results demonstrated the reliability of the statistical optimization designs for predicting the efficient removal of NGB.
Bioactive materials that can support cell adhesion and tissue regeneration are greatly in demand in clinical applications. Surface modification with bioactive molecules is an efficient strategy to convert conventional bioinert materials into bioactive materials. However, there is an urgent need to find a universal and one-step modification strategy to realize the above transformation for bioactivation. In this work, we report a universal and one-step modification strategy to easily modify and render diverse materials bioactivation by dipping materials into the solution of dibutylamine-DOPA-lysine-DOPA (DbaYKY) tripeptide-terminated cell-adhesive molecules, β-peptide polymer, or RGD peptide for only 5 min. This strategy provides materials with a stable surface modification layer and does not cause an undesired surface color change like the widely used polydopamine coating. This one-step strategy can endow material surfaces with cell adhesion properties without concerns on nonspecific conjugation of proteins and macromolecules. This universal and one-step surface bioactivation strategy implies a wide range of applications in implantable biomaterials.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)