Decidual natural killer cells dysfunction is caused by IDO downregulation in dMDSCs with Toxoplasma gondii infection.

Myeloid-derived suppressor cells (MDSCs) play a crucial role in maintaining maternal-fetal tolerance by expressing some immune-suppressive molecules, such as indoleamine 2,3-dioxygenase (IDO). Toxoplasma gondii (T. gondii) infection can break the immune microenvironment of maternal-fetal interface, resulting in adverse pregnancy outcomes. However, whether T. gondii affects IDO expression in dMDSCs and the molecular mechanism of its effect are still unclear. Here we show, the mRNA level of IDO is increased but the protein level decreased in infected dMDSCs. Mechanistically, the upregulation of transcriptional levels of IDO in dMDSCs is regulated through STAT3/p52-RelB pathway and the decrease of IDO expression is due to its degradation caused by increased SOCS3 after T. gondii infection. In vivo, the adverse pregnancy outcomes of IDO-/- infected mice are more severe than those of wide-type infected mice and obviously improved after exogenous kynurenine treatment. Also, the reduction of IDO in dMDSCs induced by T. gondii infection results in the downregulation of TGF-ß and IL-10 expression in dNK cells regulated through Kyn/AhR/SP1 signal pathway, eventually leading to the dysfunction of dNK cells and contributing the occurrence of adverse pregnancy outcomes. This study reveals a novel molecular mechanism in adverse pregnancy outcome induced by T. gondii infection.

Biomarker-triggered, spatiotemporal controlled DNA nanodevice simultaneous assembly and disassembly.

Despite many advances in the use of DNA nanodevices as assembly or disassembly modules to build various complex structures, the simultaneous assembly and disassembly of DNA structures in living cells remains a challenge. In this study, we present a modular engineering approach for assembling and disassembling DNA nanodevices in response to endogenous biomarkers. As a result of pairwise prehybridization of original DNA strands, the DNA nanodevice is initially inert. In an effort to bind one of the paired strands and release its complement, nucleolin competes. Assembly of the DNA nanodevice is initiated when the released complement binds to it, and disassembly is initiated when APE1 shears the assembled binding site of the DNA nanodevice. Spatial-temporal logic control is achieved through our approach during the assembly and disassembly of DNA nanodevices. Furthermore, by means of this assembly and disassembly procedure, the sequential detection and imaging of two tumor markers can be achieved, thereby effectively reducing false-positive signal results and accelerating the detection time. This study emphasizes the simultaneous assembly and disassembly of DNA nanodevices controlled by biomarkers in a simple and versatile manner; it has the potential to expand the application scope of DNA nanotechnology and offers an idea for the implementation of precision medicine testing.

Vascular Risk Factors and Brain Health in Aging: Insights from a Community-Based Cohort Study.

Background: The aging population and high rates of Alzheimer's disease (AD) create significant medical burdens, prompting a need for early prevention. Targeting modifiable risk factors like vascular risk factors (VRFs), closely linked to AD, may provide a promising strategy for intervention. Objective: This study investigates how VRFs influence cognitive performance and brain structures in a community-based cohort. Methods: In this cross-sectional study, 4,667 participants over 50 years old, drawn from the Beijing Ageing Brain Rejuvenation Initiative project, were meticulously examined. Cognitive function and VRFs (diabetes mellitus, hypertension, hyperlipidemia, obesity, and smoking), were comprehensively assessed through one-to-one interviews. Additionally, a subset of participants (n = 719) underwent MRI, encompassing T1-weighted and diffusion-weighted scans, to elucidate gray matter volume and white matter structural network organization. Results: The findings unveil diabetes as a potent detriment to memory, manifesting in atrophy within the right supramarginal gyrus and diminished nodal efficiency and degree centrality in the right inferior parietal lobe. Hypertension solely impaired memory without significant structural changes. Intriguingly, individuals with comorbid diabetes and hypertension exhibited the most pronounced deficits in both brain structure and cognitive performance. Remarkably, hyperlipidemia emerged as a factor associated with enhanced cognition, and preservation of brain structure. Conclusions: This study illuminates the intricate associations between VRFs and the varied patterns of cognitive and brain structural damage. Notably, the synergistic effect of diabetes and hypertension emerges as particularly deleterious. These findings underscore the imperative to tailor interventions for patients with distinct VRF comorbidities, especially when addressing cognitive decline and structural brain changes.

Exploring the effective compounds and potential mechanisms of Shengxian Decoction against coronary heart disease by UPLC-Q-TOF/MS and network pharmacology analysis.

As a well-known classical Chinese medicine prescription, Shengxian Decoction (SXD) has been applied for a century to treat cardiovascular diseases, especially coronary heart disease (CHD), but the potentially effective compounds and underlying mechanisms remain unclear. With ultra-performance liquid chromatography-quadrupole-time of flight-mass spectrometry (UPLC-Q-TOF/MS) and network pharmacology analysis, the potential effective compounds of SXD and their pharmacological mechanisms against CHD were identified and revealed. 57 effective compounds with favorable pharmacokinetic characteristics and biological activities were screened through UPLC-Q-TOF/MS analysis, database and literature mining, interacting with 96 CHD-related targets to support potential synergistic therapeutic actions. Systematic analysis of the PPI network and microarray data further revealed six core targets, including TNF, IL-1ß, IL-6, TP53, VEGFA and PTGS2, which were mainly involved in fluid shear stress and atherosclerosis, lipid and atherosclerosis, PI3K-Akt signaling pathway et al. Moreover, the proposed contribution indexes of effective compounds indicated these compounds, including isoferulic acid, quercetin, calycosin, ferulic acid, kaempferol, calycosin 7-O-glycoside, formononetin, astragaloside IV and saikosaponin D, as the core compounds of SXD. The molecular docking results confirmed that those core compound-target pairs exhibited strong binding energy. Furthermore, we validated that SXD significantly alleviated myocardial tissue injury in CHD rats and reversed H/R-induced decreases in H9c2 cell viability by attenuating the production of TNF, IL-6 and IL-1ß, and reducing cardiomyocyte apoptosis via down-regulating the TP53, caspase3 and cytochrome C mRNA expression levels as well as caspase3, caspase9 and cytochrome C protein expression levels according to RT-qPCR and Western blot results. Our findings explained the pharmacological mechanisms underlying the effectiveness of SXD in the treatment of CHD, and laid a foundation for future basic and clinical research of SXD.

Biomarker-Driven DNA-Functionalized Colloidal Programmed Simultaneous Assembly and Disassembly in Cells.

Complex structures and devices, both natural and artificial, can often undergo assembly and disassembly. Assembly and disassembly allow multiple stimuli to initiate, for example, the assembly and disassembly of primary cilia under the control of E3 ubiquitin ligases and deubiquitinases. Although biology relies on such schemes, they are rarely available in materials science. Here, we demonstrate a DNA-functionalized colloidal Au response to endogenous biomarkers to trigger simultaneous assembly and disassembly techniques. Colloidal Au is initially inert because the starting DNA strands are paired and prehybridized. TK1 mRNA competes to bind one of the paired strands and release its complement. The released complement binds to the next colloidal Au to initiate assembly, and APE1 can shear the colloidal Au assembly binding site to initiate disassembly. Our strategy provides temporal and spatial logic control during colloidal Au assembly and disassembly, and this simultaneous assembly and disassembly process can be used for sequential detection and cellular imaging of two biomarkers, effectively reducing signal false-positive results and shortening detection time. This work highlights biomarker-controlled colloidal Au simultaneous assembly and disassembly in ways that are simple and versatile, with the potential to enrich the application scope of DNA nanotechnology and provide an idea for the application of precision medicine testing.

The identification of metabolites from gut microbiota in coronary heart disease via network pharmacology.

Although the gut microbial metabolites exhibit potential effects on coronary heart disease (CHD), the underlying mechanism remains unclear. In this study, the active gut microbial metabolites acting on CHD and their potential mechanisms of action were explored through a network pharmacological approach. We collected a total of 208 metabolites from the gutMgene database and 726 overlapping targets from the similarity ensemble approach (SEA) and SwissTargetPrediction (STP) database, and ultimately identified 610 targets relevant to CHD. In conjunction with the gutMGene database, we identified 12 key targets. The targets of exogenous substances were removed, and 10 core targets involved in CHD were eventually retained. The microbiota-metabolites-targets-signalling pathways network analysis revealed that C-type lectin receptor signalling pathway, Lachnospiraceae, Escherichia, mitogen-activated protein kinase 1, prostaglandin-endoperoxidase synthase 2, phenylacetylglutamine and alcoholic acid are notable components of CHD and play important roles in the development of CHD. The results of molecular docking experiments demonstrated that AKT1-glycocholic acid and PTGS2-phenylacetylglutamine complexes may act on C-type lectin receptor signalling pathways. In this study, the key substances and potential mechanisms of gut microbial metabolites were analysed via network pharmacological methods, and a scientific basis and comprehensive idea were provided for the effects of gut microbial metabolites on CHD.

Tailoring Classical Conditioning Behavior in TiO2 Nanowires: ZnO QDs-Based Optoelectronic Memristors for Neuromorphic Hardware.

Neuromorphic hardware equipped with associative learning capabilities presents fascinating applications in the next generation of artificial intelligence. However, research into synaptic devices exhibiting complex associative learning behaviors is still nascent. Here, an optoelectronic memristor based on Ag/TiO2 Nanowires: ZnO Quantum dots/FTO was proposed and constructed to emulate the biological associative learning behaviors. Effective implementation of synaptic behaviors, including long and short-term plasticity, and learning-forgetting-relearning behaviors, were achieved in the device through the application of light and electrical stimuli. Leveraging the optoelectronic co-modulated characteristics, a simulation of neuromorphic computing was conducted, resulting in a handwriting digit recognition accuracy of 88.9%. Furthermore, a 3 × 7 memristor array was constructed, confirming its application in artificial visual memory. Most importantly, complex biological associative learning behaviors were emulated by mapping the light and electrical stimuli into conditioned and unconditioned stimuli, respectively. After training through associative pairs, reflexes could be triggered solely using light stimuli. Comprehensively, under specific optoelectronic signal applications, the four features of classical conditioning, namely acquisition, extinction, recovery, and generalization, were elegantly emulated. This work provides an optoelectronic memristor with associative behavior capabilities, offering a pathway for advancing brain-machine interfaces, autonomous robots, and machine self-learning in the future.

Micropyramid Array Bimodal Electronic Skin for Intelligent Material and Surface Shape Perception Based on Capacitive Sensing.

Developing electronic skins (e-skins) that are comparable to or even beyond human tactile perception holds significant importance in advancing the process of intellectualization. In this context, a machine-learning-motivated micropyramid array bimodal (MAB) e-skin based on capacitive sensing is reported, which enables spatial mapping applications based on bimodal sensing (proximity and pressure) implemented via fringing and iontronic effects, such as contactless measurement of 3D objects and contact recognition of Braille letters. Benefiting from the iontronic effect and single-micropyramid structure, the MAB e-skin in pressure mode yields impressive features: a maximum sensitivity of 655.3 kPa-1 (below 0.5 kPa), a linear sensitivity of 327.9 kPa-1 (0.5-15 kPa), and an ultralow limit of detection of 0.2 Pa. With the assistance of multilayer perceptron and convolutional neural network, the MAB e-skin can accurately perceive 6 materials and 10 surface shapes based on the training and learning using the collected datasets from proximity and pressure modes, thus allowing it to achieve the precise perception of different objects within one proximity-pressure cycle. The development of this MAB e-skin opens a new avenue for robotic skin and the expansion of advanced applications.

CrossFormer++: A Versatile Vision Transformer Hinging on Cross-Scale Attention.

While features of different scales are perceptually important to visual inputs, existing vision transformers do not yet take advantage of them explicitly. To this end, we first propose a cross-scale vision transformer, CrossFormer. It introduces a cross-scale embedding layer (CEL) and a long-short distance attention (LSDA). On the one hand, CEL blends each token with multiple patches of different scales, providing the self-attention module itself with cross-scale features. On the other hand, LSDA splits the self-attention module into a short-distance one and a long-distance counterpart, which not only reduces the computational burden but also keeps both small-scale and large-scale features in the tokens. Moreover, through experiments on CrossFormer, we observe another two issues that affect vision transformers' performance, i.e., the enlarging self-attention maps and amplitude explosion. Thus, we further propose a progressive group size (PGS) paradigm and an amplitude cooling layer (ACL) to alleviate the two issues, respectively. The CrossFormer incorporating with PGS and ACL is called CrossFormer++. Extensive experiments show that CrossFormer++ outperforms the other vision transformers on image classification, object detection, instance segmentation, and semantic segmentation tasks.

OBMO: One Bounding Box Multiple Objects for Monocular 3D Object Detection.

Compared to typical multi-sensor systems, monocular 3D object detection has attracted much attention due to its simple configuration. However, there is still a significant gap between LiDAR-based and monocular-based methods. In this paper, we find that the ill-posed nature of monocular imagery can lead to depth ambiguity. Specifically, objects with different depths can appear with the same bounding boxes and similar visual features in the 2D image. Unfortunately, the network cannot accurately distinguish different depths from such non-discriminative visual features, resulting in unstable depth training. To facilitate depth learning, we propose a simple yet effective plug-and-play module, One Bounding Box Multiple Objects (OBMO). Concretely, we add a set of suitable pseudo labels by shifting the 3D bounding box along the viewing frustum. To constrain the pseudo-3D labels to be reasonable, we carefully design two label scoring strategies to represent their quality. In contrast to the original hard depth labels, such soft pseudo labels with quality scores allow the network to learn a reasonable depth range, boosting training stability and thus improving final performance. Extensive experiments on KITTI and Waymo benchmarks show that our method significantly improves state-of-the-art monocular 3D detectors by a significant margin (The improvements under the moderate setting on KITTI validation set are 1.82 ~ 10.91% mAP in BEV and 1.18 ~ 9.36% mAP in 3D). Codes have been released at https://github.com/mrsempress/OBMO.

Differential regulation of hair cell actin cytoskeleton mediated by SRF and MRTFB.

The MRTF-SRF pathway has been extensively studied for its crucial role in driving the expression of a large number of genes involved in actin cytoskeleton of various cell types. However, the specific contribution of MRTF-SRF in hair cells remains unknown. In this study, we showed that hair cell-specific deletion of Srf or Mrtfb, but not Mrtfa, leads to similar defects in the development of stereocilia dimensions and the maintenance of cuticular plate integrity. We used fluorescence-activated cell sorting-based hair cell RNA-Seq analysis to investigate the mechanistic underpinnings of the changes observed in Srf and Mrtfb mutants, respectively. Interestingly, the transcriptome analysis revealed distinct profiles of genes regulated by Srf and Mrtfb, suggesting different transcriptional regulation mechanisms of actin cytoskeleton activities mediated by Srf and Mrtfb. Exogenous delivery of calponin 2 using Adeno-associated virus transduction in Srf mutants partially rescued the impairments of stereocilia dimensions and the F-actin intensity of cuticular plate, suggesting the involvement of Cnn2, as an Srf downstream target, in regulating the hair bundle morphology and cuticular plate actin cytoskeleton organization. Our study uncovers, for the first time, the unexpected differential transcriptional regulation of actin cytoskeleton mediated by Srf and Mrtfb in hair cells, and also demonstrates the critical role of SRF-CNN2 in modulating actin dynamics of the stereocilia and cuticular plate, providing new insights into the molecular mechanism underlying hair cell development and maintenance.

Biomolecular Markers of Brain Aging.

Characterized by the gradual loss of physiological integrity, impaired function, and increased susceptibility to death, aging is considered the primary risk factor for major human diseases, such as cancer, diabetes, cardiovascular disorders, and neurodegenerative diseases. The time-dependent accumulation of cellular damage is widely considered the general cause of aging. While the mechanism of normal aging is still unresolved, researchers have identified different markers of aging, including genomic instability, telomere attrition, epigenetic alterations, loss of proteostasis, deregulated nutrient-sensing, mitochondrial dysfunction, cellular senescence, stem cell exhaustion, and altered intercellular communication. Theories of aging can be divided into two categories: (1) aging is a genetically programmed process, and (2) aging is a random process caused by gradual damage to the organism over time as a result of its vital activities. Aging affects the entire human body, and aging of the brain is undoubtedly different from all other organs, as neurons are highly differentiated postmitotic cells, and the lifespan of most neurons in the postnatal period is equal to the lifespan of the brain. In this chapter, we discuss the conserved mechanisms of aging that may underlie the changes observed in the aging brain, with a focus on mitochondrial function and oxidative stress, autophagy and protein turnover, insulin/IGF signaling, target of rapamycin (TOR) signaling, and sirtuin function.

White matter changes underlie hypertension-related cognitive decline in older adults.

Hypertension has been well recognized as a risk factor for cognitive impairment and dementia. Although the underlying mechanisms of hypertension-affected cognitive deterioration are not fully understood, white matter changes (WMCs) seem to play an important role. WMCs include low microstructural integrity and subsequent white matter macrostructural lesions, which are common on brain imaging in hypertensive patients and are critical for multiple cognitive domains. This article provides an overview of the impact of hypertension on white matter microstructural and macrostructural changes and its link to cognitive dysfunction. Hypertension may induce microstructural changes in white matter, especially for the long-range fibers such as anterior thalamic radiation (ATR) and inferior fronto-occipital fasciculus (IFOF), and then macrostructural abnormalities affecting different lobes, especially the periventricular area. Different regions' WMCs would further exert different effects to specific cognitive domains and accelerate brain aging. As a modifiable risk factor, hypertension might provide a new perspective for alleviating and delaying cognitive impairment.

Role of herbal medicine and gut microbiota in the prevention and treatment of obesity.

ETHNOPHARMACOLOGICAL RELEVANCE: Obesity is a common metabolic dysfunction disease, which is highly correlated with the homeostasis of gut microbiota (GM). The dysregulation of GM on energy metabolism, immune response, insulin resistance and endogenous metabolites (e.g., short chain fatty acids and secondary bile acids) can affect the occurrence and development of obesity. Herbal medicine (HM) has particular advantages and definite therapeutic effects in the prevention and treatment of obesity, but its underlying mechanism is not fully clear. AIM OF THE STUDY: In this review, the representative basic and clinical anti-obesity studies associated with the homeostasis of GM regulated by HM including active components, single herb and herbal formulae were summarized and discussed. We aim to provide a state of art reference for the mechanism research of HM in treating obesity and the further development of new anti-obesity drugs. MATERIALS AND METHODS: The relevant information was collected by searching keywords (obesity, herbal medicine, prescriptions, mechanism, GM, short chain fatty acids, etc.) from scientific databases (CNKI, PubMed, SpringerLink, Web of Science, SciFinder, etc.). RESULTS: GM dysbiosis did occur in obese patients and mice, whiles the intervention of GM could ameliorate the condition of obesity. HM (e.g., berberine, Ephedra sinica, Rehjnannia glutinosa, and Buzhong Yiqi prescription) has been proved to possess a certain regulation on GM and an explicit effect on obesity, but the exact mechanism of HM in improving obesity by regulating GM remains superficial. CONCLUSION: GM is involved in HM against obesity, and GM can be a novel therapeutic target for treating obesity.

Anti-obesity and Gut Microbiota Modulation Effect of Astragalus Polysaccharides Combined with Berberine on High-Fat Diet-Fed Obese Mice.

OBJECTIVE: To investigate whether astragalus polysaccharides (APS) combined with berberine (BBR) can reduce high-fat diet (HFD)-induced obesity in mice. METHODS: Except for normal mice, 32 HFD-induced obese mice were randomized into HFD, APS (1,000 mg/kg APS), BBR (200 mg/kg BBR), and APS plus BBR (1,000 mg/kg APS plus 200 mg/kg BBR) groups, respectively. After 6-week treatment (once daily by gavage), the obesity phenotype and pharmacodynamic effects were evaluated by histopathological examination of epididymal fat, liver, and colon using hematoxylin-eosin staining and serum biochemical analyses by an automated chemistry analyzer. The feces were collected at the 12 th week, and taxonomic and functional profiles of gut microbiota were analyzed by 16S ribosomal ribonucleic acid (16S rRNA) sequencing. RESULTS: Compared with HFD group, the average body weight of APS plus BBR group was decreased (P<0.01), accompanied with the reduced fat accumulation, enhanced colonic integrity, insulin sensitivity and glucose homeostasis (P<0.05 or P<0.01). Importantly, APS combined with BBR treatment was more effective than APS or BBR alone in improving HFD-induced insulin resistance (P<0.05 or P<0.01). 16S rRNA sequence-based analysis of fecal samples demonstrated that APS combined with BBR treatment exhibited a better impact on HFD-induced gut microbiota dysbiosis, exclusively via the enriched abundances of Bacteroides, which corresponded to the large increase of predicted bacterial genes involved in carbohydrate metabolism. CONCLUSION: APS combined with BBR may synergistically reduce obesity and modulate the gut microbiota in HFD-fed mice.

CST1 promotes the proliferation and migration of PDGF-BB-treated airway smooth muscle cells via the PI3K/AKT signaling pathway.

Typically, airway remodeling caused by migration and proliferation of airway smooth muscle cells (ASMCs) plays a crucial role in the pathophysiological characteristics of asthma development. Cystatin 1 (CST1), a protein-coding gene referred to as Cystatin SN, is highly expressed in asthma patients. However, the role of CST1 and related molecular mechanisms in the development of asthma remains to be explored. This study aims to investigate the role of CST1 in asthma progression and present related molecular mechanisms. To explore these aspects, human ASMCs with platelet-derived growth factor BB (PDGF-BB) are initially stimulated and applied as a cellular model of asthma. Further, CST1 is knocked down with small interfering ribose nucleic acid (siRNA) overexpressed with plasmids. Then, 5-ethynyl-2'-deoxyuridine (EdU) and Cell Count Kit (CCK)-8 assays are applied to assess the cell proliferation rates. Further, Transwell and Western blot analyses for migration of cells and expression of MMP1 and MMP9 proteins are assessed, respectively. Under PDGF-BB stimulation, human ASMCs showed an increased CST1 expression, enhanced proliferation, and migration abilities, as well as up-regulated PI3K/AKT signaling pathway. Further, knockdown or overexpression of CST1 presented the declined or enhanced proliferation, migration, and up-regulation of the PI3K/AKT signaling pathway of human ASMCs. Inhibiting PI3K/AKT signaling pathway displayed the reduced migration and proliferation of human ASMCs. In summary, these findings indicated that CST1 played an essential role in the progression of asthma by activating the PI3K/AKT signaling pathway and promoting the migration and proliferation abilities of human ASMCs treated with PDGF-BB.

New Alloy of an Al-Chalcogen System: AlSe Surface Alloys on Al(111).

Metal chalcogenides are a promising material for novel physical research and nanoelectronic device applications. Here, we systematically investigate the crystal structure and electronic properties of AlSe alloys on Al(111) using scanning tunneling microscopy, angle-resolved photoelectron spectrometry, and first-principle calculations. We reveal that the AlSe surface alloy possesses a closed-packed atomic structure. The AlSe surface alloy comprises two atomic sublayers (Se sublayer and Al sublayer) with a height difference of 1.16 Å. Our results indicate that the AlSe alloy hosts two hole-like bands, which are mainly derived from the in-plane orbital of AlSe (p x and p y ). These two bands located at about -2.22 ±0.01 eV around the Gamma point, far below the Fermi level, distinguished from other metal chalcogenides and binary alloys. AlSe alloys have the advantages of large-scale atomic flat terraces and a wide band gap, appropriate to serve as an interface layer for two-dimensional materials. Meanwhile, our results provide implications for related Al-chalcogen interfaces.

Methodology of network pharmacology for research on Chinese herbal medicine against COVID-19: A review.

Traditional Chinese medicine, as a complementary and alternative medicine, has been practiced for thousands of years in China and possesses remarkable clinical efficacy. Thus, systematic analysis and examination of the mechanistic links between Chinese herbal medicine (CHM) and the complex human body can benefit contemporary understandings by carrying out qualitative and quantitative analysis. With increasing attention, the approach of network pharmacology has begun to unveil the mystery of CHM by constructing the heterogeneous network relationship of "herb-compound-target-pathway," which corresponds to the holistic mechanisms of CHM. By integrating computational techniques into network pharmacology, the efficiency and accuracy of active compound screening and target fishing have been improved at an unprecedented pace. This review dissects the core innovations to the network pharmacology approach that were developed in the years since 2015 and highlights how this tool has been applied to understanding the coronavirus disease 2019 and refining the clinical use of CHM to combat it.

Toward Better Accuracy-Efficiency Trade-Offs: Divide and Co-Training.

The width of a neural network matters since increasing the width will necessarily increase the model capacity. However, the performance of a network does not improve linearly with the width and soon gets saturated. In this case, we argue that increasing the number of networks (ensemble) can achieve better accuracy-efficiency trade-offs than purely increasing the width. To prove it, one large network is divided into several small ones regarding its parameters and regularization components. Each of these small networks has a fraction of the original one's parameters. We then train these small networks together and make them see various views of the same data to increase their diversity. During this co-training process, networks can also learn from each other. As a result, small networks can achieve better ensemble performance than the large one with few or no extra parameters or FLOPs, i. e., achieving better accuracy-efficiency trade-offs. Small networks can also achieve faster inference speed than the large one by concurrent running. All of the above shows that the number of networks is a new dimension of model scaling. We validate our argument with 8 different neural architectures on common benchmarks through extensive experiments.

Head-to-Head Comparison of Different Blood Collecting Tubes for Quantification of Alzheimer's Disease Biomarkers in Plasma.

To examine whether the type of blood collection tubes affects the quantification of plasma biomarkers for Alzheimer's disease analyzed with a single-molecule array (Simoa), we recruited a healthy cohort (n = 34, 11 males, mean age = 28.7 ± 7.55) and collected plasma in the following tubes: dipotassium ethylenediaminetetraacetic acid (K2-EDTA), heparin lithium (Li-Hep), and heparin sodium (Na-Hep). Plasma tau, phosphorylated tau 181 (p-tau181), amyloid ß (1-40) (Aß40), and amyloid ß (1-42) (Aß42) were quantified using Simoa. We compared the value of plasma analytes, as well as the effects of sex on the measurements. We found that plasma collected in Li-Hep and Na-Hep tubes yielded significantly higher tau and p-tau181 levels compared to plasma collected in K2-EDTA tubes from the same person, but there was no difference in the measured values of the Aß40, Aß42, and Aß42/40 ratio. Therefore, the type of blood collecting tubes should be considered when planning studies that measure plasma tau.