Retrosynthesis is a procedure where target molecule transformed into potential reactants and thus the synthesis routes can be identified. Recently, computational approaches have been developed to accelerate design of routes. In this paper, we develop generative framework $\mathsf{G^2Retro}$ for one-step retrosynthesis prediction. imitates reversed logic synthetic reactions. It first predicts reaction centers in molecules (products), identifies synthons needed assemble products, transforms...

The spleen emerges as a pivotal target for mRNA delivery, prompting continual quest specialized and efficient lipid nanoparticles (LNPs) designed to enhance spleen-selective transfection efficiency. Here we report imidazole-containing ionizable lipids (IMILs) that demonstrate pronounced preference delivery into the with exceptional We optimized IMIL structures by constructing screening multidimensional library containing multiple heads, tails, linkers perform structure–activity correlation...

Cation substitution is a straightforward but effective technique for improving the structure and properties; however, controlling directed still poses significant difficulties. Herein, metal-free hydroxyfluorooxoborate (NH4)[C(NH2)3][B3O3F4(OH)] has been synthesized using strategy of heterologous based on template A2[B3O3F4(OH)]. Tunable optical properties have achieved via varied A-site cation substitution. The intrinsic mechanism this tunability was established by crystallography...

Drug development is a critical but notoriously resource- and time-consuming process. In this manuscript, we develop novel generative artificial intelligence (genAI) method DiffSMol to facilitate drug development. DiffSmol generates 3D binding molecules based on the shapes of known ligands. encapsulates geometric details ligand within pre-trained, expressive shape embeddings then new through diffusion model. further modifies generated structures iteratively via guidance better resemble...

MHC Class I protein plays an important role in immunotherapy by presenting immunogenic peptides to anti-tumor immune cells. The repertoires of for various proteins are distinct, which can be reflected their diverse binding motifs. To characterize motifs proteins, vitro experiments have been conducted screen with high affinities hundreds given proteins. However, considering tens thousands known conducting extensive is infeasible, and thus a more efficient scalable way needed.

Self-attention (SA) mechanisms have been widely used in developing sequential recommendation (SR) methods, and demonstrated state-of-the-art performance. However, this article, we show that self-attentive SR methods substantially suffer from the over-smoothing issue item embeddings within a sequence become increasingly similar across attention blocks. As literature, could lead to loss of information individual items, significantly degrade models’ scalability To address issue, view items...

Mature microRNAs (miRNAs) are short, single-stranded RNAs that bind to target mRNAs and induce translational repression gene silencing. Many miRNAs discovered in animals have been implicated diseases recently pursued as therapeutic targets. However, conventional pharmacological screening for candidate small-molecule drugs can be time-consuming labor-intensive. Therefore, developing a computational program assist mature miRNA-targeted drug discovery silico is desirable. Our previous work...

Effective and successful clinical trials are essential in developing new drugs advancing treatments. However, very expensive easy to fail. The high cost low success rate of motivate research on inferring knowledge from existing innovative ways for designing future trials. In this manuscript, we present our efforts constructing the first publicly available Clinical Trials Knowledge Graph, denoted as [Formula: see text]. text] includes nodes representing medical entities (e.g., studies,...

T-cell receptors can recognize foreign peptides bound to major histocompatibility complex (MHC) class-I proteins, and thus trigger the adaptive immune response. Therefore, identifying that bind MHC molecules plays a vital role in design of peptide vaccines. Many computational methods, for example, state-of-the-art allele-specific method <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" id="m1"><mml:mrow><mml:mi mathvariant="monospace">MHCflurry</mml:mi></mml:mrow></mml:math> , have...

Changes in slope aspect have important effects on the C, N and P nutrient contents stoichiometric characteristics of plants soils. As an forest type subtropical region, Pinus massoniana forests play role restoration development ecosystems. In this study, effect stoichiometry leaves soils four P. types (i.e., pure (PF), massoniana-Liquidambar formosana mixed (PLM), - Platycarya strobilacea (PPM), massoniana-Quercus acutissima (PQM)) north region China were studied. The results showed that had...

Ligand-based drug design aims to identify novel candidates of similar shapes with known active molecules. In this paper, we formulated an in silico shape-conditioned molecule generation problem generate 3D structures conditioned on the shape a given molecule. To address problem, developed translation- and rotation-equivariant shape-guided generative model ShapeMol. ShapeMol consists equivariant encoder that maps molecular surface into latent embeddings, diffusion generates molecules based...

KNN (K nearest neighbor) algorithm is a widely used regression method, with very simple principle about neighborhood. Though it achieves success in many application areas, the method has shortcoming of weighting equal contributions to each attribute when computing distance between instances. In this paper, we applied weighted approach by using weights obtained from optimization and feature selection methods compared performance efficiency these two types algorithms problems. Experiments on...

In this work, we investigated the effect of ${\rm^{15}N(p,\alpha)^{12}C}$ reaction produced by collision between proton and ammonia monohydrate on kinetic energy release (KER) water molecule fragmentation. After occurrence nuclear reaction, it was found that charge states $q$ flight speeds $v$ are main factors affecting KER With value $q/v$ increases, distribution gets wider peak position changes more pronounced. The gained each fragment is related to mass distance from reaction. study,...

The emergence of Poly (ADP-ribose) polymerase inhibitors (PARPi) has marked the beginning a precise targeted therapy era for ovarian cancer.However, an increasing number patients are experiencing primary or acquired resistance to PARPi, severely limiting its clinical application.Deciphering underlying mechanisms PARPi and discovering new therapeutic targets is urgent critical issue address.In this study, we observed close correlation between glycolysis, tumor angiogenesis, in...

This work presents a novel RGB-D dynamic simultaneous localization and mapping (SLAM) method that improves accuracy, stability, efficiency of while relying on deep learning in environment, contrast to traditional static scene-based visual SLAM methods. Based the classic framework SLAM, we propose replaces feature extraction with convolutional neural network approach, aiming enhance accuracy localization, as well improve algorithm’s ability capture represent characteristics entire scene....

ABSTRACT Effective and successful clinical trials are essential in developing new drugs advancing treatments. However, very expensive easy to fail. The high cost low success rate of motivate research on inferring knowledge from existing innovative ways for designing future trials. In this manuscript, we present our efforts constructing the first publicly available Clinical Trials Knowledge Graph, denoted as CTKG . includes nodes representing medical entities (e.g., studies, conditions),...