MIFAM-DTI: a drug-target interactions predicting model based on multi-source information fusion and attention mechanism.

Accurate identification of potential drug-target pairs is a crucial step in drug development and drug repositioning, which is characterized by the ability of the drug to bind to and modulate the activity of the target molecule, resulting in the desired therapeutic effect. As machine learning and deep learning technologies advance, an increasing number of models are being engaged for the prediction of drug-target interactions. However, there is still a great challenge to improve the accuracy and efficiency of predicting. In this study, we proposed a deep learning method called Multi-source Information Fusion and Attention Mechanism for Drug-Target Interaction (MIFAM-DTI) to predict drug-target interactions. Firstly, the physicochemical property feature vector and the Molecular ACCess System molecular fingerprint feature vector of a drug were extracted based on its SMILES sequence. The dipeptide composition feature vector and the Evolutionary Scale Modeling -1b feature vector of a target were constructed based on its amino acid sequence information. Secondly, the PCA method was employed to reduce the dimensionality of the four feature vectors, and the adjacency matrices were constructed by calculating the cosine similarity. Thirdly, the two feature vectors of each drug were concatenated and the two adjacency matrices were subjected to a logical OR operation. And then they were fed into a model composed of graph attention network and multi-head self-attention to obtain the final drug feature vectors. With the same method, the final target feature vectors were obtained. Finally, these final feature vectors were concatenated, which served as the input to a fully connected layer, resulting in the prediction output. MIFAM-DTI not only integrated multi-source information to capture the drug and target features more comprehensively, but also utilized the graph attention network and multi-head self-attention to autonomously learn attention weights and more comprehensively capture information in sequence data. Experimental results demonstrated that MIFAM-DTI outperformed state-of-the-art methods in terms of AUC and AUPR. Case study results of coenzymes involved in cellular energy metabolism also demonstrated the effectiveness and practicality of MIFAM-DTI. The source code and experimental data for MIFAM-DTI are available at https://github.com/Search-AB/MIFAM-DTI.

Senescence Promotes the Recovery of Stemness among Cancer Cells via Reprograming.

Both the senescence of cancer cells and the maintenance of cancer stem cells seem to be mutually exclusive because senescence is considered a physiological mechanism that effectively suppresses tumor growth. Recent studies have revealed common signaling pathways between cellular senescence and the maintenance of stemness in cancer cells, thus challenging the conventional understanding of this process. Although the links between these processes have not yet been fully elucidated, emerging evidence indicates that senescent cancer cells can undergo reprograming to recover stemness. Herein, we provide a comprehensive overview of the close correlation between senescence and stemness reprograming in cancer cells, with a particular focus on the mechanisms by which senescent cancer cells recover their stemness in various tumor systems.

RAB27B-regulated exosomes mediate LSC maintenance via resistance to senescence and crosstalk with the microenvironment.

The fate of leukaemia stem cells (LSCs) is determined by both their inherent mechanisms and crosstalk with their niches. Although LSCs were confirmed to be eradicated by restarting senescence, the specific key regulators of LSC resistance to senescence and remodelling of the niche to obtain a microenvironment suitable for stemness remain unknown. Here, we found that RAB27B, a gene regulating exosome secretion, was overexpressed in LSCs and associated with the poor prognosis of acute myeloid leukaemia (AML) patients. The increased RAB27B in LSCs prevented their senescence and maintained their stemness in vitro and in vivo. Mechanically, the increased RAB27B expression in LSCs selectively promoted the loading and release of exosomes rich in senescence-inducing proteins by direct combination. Furthermore, RAB27B-regulated LSC-derived exosomes remodelled the niche and induced senescence of mesenchymal stem cells (MSCs) with increased RAB27B expression ex vivo and in vivo. The increased RAB27B in the senescent MSCs conversely promoted LSC maintenance ex vivo and in vivo via selective excretion of exosomes rich in stemness-promoting proteins. Therefore, we identified the specifically increased RAB27B in LSCs and their educated senescent MSCs as a hub molecule for LSC resistance to senescence and maintenance through crosstalk with its niche via selective exosome excretion.

Road Network-Guided Fine-Grained Urban Traffic Flow Inference.

Accurate inference of fine-grained traffic flow from coarse-grained one is an emerging yet crucial problem, which can help greatly reduce the number of the required traffic monitoring sensors for cost savings. In this work, we note that traffic flow has a high correlation with road network, which was either completely ignored or simply treated as an external factor in previous works. To facilitate this problem, we propose a novel road-aware traffic flow magnifier (RATFM) that explicitly exploits the prior knowledge of road networks to fully learn the road-aware spatial distribution of fine-grained traffic flow. Specifically, a multidirectional 1-D convolutional layer is first introduced to extract the semantic feature of the road network. Subsequently, we incorporate the road network feature and coarse-grained flow feature to regularize the short-range spatial distribution modeling of road-relative traffic flow. Furthermore, we take the road network feature as a query to capture the long-range spatial distribution of traffic flow with a transformer architecture. Benefiting from the road-aware inference mechanism, our method can generate high-quality fine-grained traffic flow maps. Extensive experiments on three real-world datasets show that the proposed RATFM outperforms state-of-the-art models under various scenarios. Our code and datasets are released at https://github.com/luimoli/RATFM.

Engineered mesenchymal stem cell exosomes loaded with miR-34c-5p selectively promote eradication of acute myeloid leukemia stem cells.

Most patients with acute myeloid leukemia (AML) relapse eventually because of the inability to effectively eliminate leukemia stem cells (LSCs), prompting the search of new therapies to eradicate LSCs. Our previous study demonstrated that miR-34c-5p promotes the clearance of LSCs in an AML mouse model, highlighting its potential as a therapeutic target for eradicating LSCs, but the effective delivery of miR-34c-5p to LSCs remains a great challenge. Here, we employed simultaneous two-step modifications to engineer mesenchymal stem cells (MSCs) and MSC-derived exosomes to create exosomes overexpressing the fused protein lysosome-associated membrane protein 2-interleukin 3 (Lamp2b-IL3) and hematopoietic cell E-selectin/L-selectin ligand (HCELL), and demonstrated that the engineered exosomes exhibited an enhanced ability for bone marrow homing and selective targeting of LSCs. Additionally, using a humanized AML mouse model, we confirmed that the engineered exosomes, loaded with miR-34c-5p, could selectively promote eradication of LSCs and impede the AML development in vivo. In summary, we successfully designed an effective delivery system and provided new insights into the development of novel therapies for delivering miRNA or other molecules to LSCs with greater cellular targeting specificity.

Refining 2SVCA Method for Measuring Telehealth Accessibility of Primary Care Physicians in Baton Rouge, Louisiana.

Equity in health care delivery is a longstanding concern of public health policy. Telehealth is considered an important way to level the playing field by broadening health services access and improving quality of care and health outcomes. This study refines the recently developed "2-Step Virtual Catchment Area (2SVCA) method" to assess the telehealth accessibility of primary care in the Baton Rouge Metropolitan Statistical Area, Louisiana. The result is compared to that of spatial accessibility via physical visits to care providers based on the popular 2-Step Floating Catchment Area (2SFCA) method. The study shows that both spatial and telehealth accessibilities decline from urban to low-density and then rural areas. Moreover, disproportionally higher percentages of African Americans are in areas with higher spatial accessibility scores; but such an advantage is not realized in telehealth accessibility. In the study area, absence of broadband availability is mainly a rural problem and leads to a lower average telehealth accessibility than physical accessibility in rural areas. On the other side, lack of broadband affordability is a challenge across the rural-urban continuum and is disproportionally associated with high concentrations of disadvantaged population groups such as households under the poverty level and Blacks.

Disparities in spatial accessibility of primary care in Louisiana: From physical to virtual accessibility.

Telehealth has been widely employed and has transformed how healthcare is delivered in the United States as a result of COVID-19 pandemic. While telehealth is utilized and encouraged to reduce the cost and travel burden for access to healthcare, there are debates on whether telehealth can promote equity in healthcare services by narrowing the gap among diverse groups. Using the Two-Step Floating Catchment Area (2SFCA) and Two-Step Virtual Catchment Area (2SVCA) methods, this study compares the disparities of physical and virtual access to primary care physicians (PCPs) in Louisiana. Both physical and virtual access to PCPs exhibit similar spatial patterns with higher scores concentrated in urban areas, followed by low-density and rural areas. However, the two accessibility measures diverge where broadband availability and affordability come to play an important role. Residents in rural areas experience additive disadvantage of even more limited telehealth accessibility than physical accessibility due to lack of broadband service provision. Areas with greater Black population proportions tend to have better physical accessibility, but such an advantage is eradicated for telehealth accessibility because of lower broadband subscription rates in these neighborhoods. Both physical and virtual accessibility scores decline in neighborhoods with higher Area Deprivation Index (ADI) values, and the disparity is further widened for in virtual accessibility compared to than physical accessibility. The study also examines how factors such as urbanicity, Black population proportion, and ADI interact in their effects on disparities of the two accessibility measures.

The role of exosomes in the stemness maintenance and progression of acute myeloid leukemia.

Acute myeloid leukemia (AML) is an aggressive malignancy of myeloid hematopoietic cells, which is characterized by the aberrant clonal proliferation of immature myeloblasts and compromised hematopoiesis. The leukemic cell population is strongly heterogeneous. Leukemic stem cells (LSCs) are an important leukemic cell subset with stemness characteristics and self-renewal ability, which contribute to the development of refractory or relapsed AML. It is now acknowledged that LSCs develop from hematopoietic stem cells (HSCs) or phenotypically directed cell populations with transcriptional stemness characteristics under selective pressure from the bone marrow (BM) niche. Exosomes are extracellular vesicles containing bioactive substances involved in intercellular communication and material exchange under steady state and pathological conditions. Several studies have reported that exosomes mediate molecular crosstalk between LSCs, leukemic blasts, and stromal cells in the BM niche, promoting LSC maintenance and AML progression. This review briefly describes the process of LSC transformation and the biogenesis of exosomes, highlighting the role of leukemic-cell- and BM-niche-derived exosomes in the maintenance of LSCs and AML progression. In addition, we discuss the potential application of exosomes in the clinic as biomarkers, therapeutic targets, and carriers for targeted drug delivery.

Development of Nafion/single-walled carbon nanotube integrated arrays for the rapid detection of salbutamol doping.

Salbutamol (SAL), a drug originally intended for the treatment of bronchial and pulmonary diseases, has repeatedly been used for doping in competitive sports. Herein, an integrated array (NFCNT array) that prepared by template-assisted scalable filtration using Nafion-coated single-walled carbon nanotube (SWCNT) is presented for the rapid field detection of SAL. Spectroscopic and microscopic measurements were used to confirm the introduction of Nafion onto the surface of the array and to analyze the resulting morphological changes. The effects of Nafion addition on the resistance and electrochemical properties of the arrays (e.g., the electrochemically active area, charge-transfer resistance, and adsorption charge) are also discussed in depth. With an electrolyte/Nafion/SWCNT interface and moderate resistance, the NFCNT-4 array prepared containing 0.04 wt% Nafion suspension exhibits the greatest voltammetric response to SAL. Subsequently, a possible mechanism for the oxidation of SAL was proposed, and a calibration curve in the range of 0.1-15 µM was established. Finally, the NFCNT-4 arrays were applied to the detection of SAL in human urine samples with satisfactory recoveries.

Factors and Methods for the Detection of Gene Expression Regulation.

Gene-expression regulation involves multiple processes and a range of regulatory factors. In this review, we describe the key factors that regulate gene expression, including transcription factors (TFs), chromatin accessibility, histone modifications, DNA methylation, and RNA modifications. In addition, we also describe methods that can be used to detect these regulatory factors.

Aerial Images Meet Crowdsourced Trajectories: A New Approach to Robust Road Extraction.

Land remote-sensing analysis is a crucial research in earth science. In this work, we focus on a challenging task of land analysis, i.e., automatic extraction of traffic roads from remote-sensing data, which has widespread applications in urban development and expansion estimation. Nevertheless, conventional methods either only utilized the limited information of aerial images, or simply fused multimodal information (e.g., vehicle trajectories), thus cannot well recognize unconstrained roads. To facilitate this problem, we introduce a novel neural network framework termed cross-modal message propagation network (CMMPNet), which fully benefits the complementary different modal data (i.e., aerial images and crowdsourced trajectories). Specifically, CMMPNet is composed of two deep autoencoders for modality-specific representation learning and a tailor-designed dual enhancement module for cross-modal representation refinement. In particular, the complementary information of each modality is comprehensively extracted and dynamically propagated to enhance the representation of another modality. Extensive experiments on three real-world benchmarks demonstrate the effectiveness of our CMMPNet for robust road extraction benefiting from blending different modal data, either using image and trajectory data or image and light detection and ranging (LiDAR) data. From the experimental results, we observe that the proposed approach outperforms current state-of-the-art methods by large margins. Our source code is resealed on the project page http://lingboliu.com/multimodal_road_extraction.html.

Online Metro Origin-Destination Prediction via Heterogeneous Information Aggregation.

Metro origin-destination prediction is a crucial yet challenging time-series analysis task in intelligent transportation systems, which aims to accurately forecast two specific types of cross-station ridership, i.e., Origin-Destination (OD) one and Destination-Origin (DO) one. However, complete OD matrices of previous time intervals can not be obtained immediately in online metro systems, and conventional methods only used limited information to forecast the future OD and DO ridership separately. In this work, we proposed a novel neural network module termed Heterogeneous Information Aggregation Machine (HIAM), which fully exploits heterogeneous information of historical data (e.g., incomplete OD matrices, unfinished order vectors, and DO matrices) to jointly learn the evolutionary patterns of OD and DO ridership. Specifically, an OD modeling branch estimates the potential destinations of unfinished orders explicitly to complement the information of incomplete OD matrices, while a DO modeling branch takes DO matrices as input to capture the spatial-temporal distribution of DO ridership. Moreover, a Dual Information Transformer is introduced to propagate the mutual information among OD features and DO features for modeling the OD-DO causality and correlation. Based on the proposed HIAM, we develop a unified Seq2Seq network to forecast the future OD and DO ridership simultaneously. Extensive experiments conducted on two large-scale benchmarks demonstrate the effectiveness of our method for online metro origin-destination prediction. Our code is resealed at https://github.com/HCPLab-SYSU/HIAM.

A non-destructive coconut fruit and seed traits extraction method based on Micro-CT and deeplabV3+ model.

With the completion of the coconut gene map and the gradual improvement of related molecular biology tools, molecular marker-assisted breeding of coconut has become the next focus of coconut breeding, and accurate coconut phenotypic traits measurement will provide technical support for screening and identifying the correspondence between genotype and phenotype. A Micro-CT system was developed to measure coconut fruits and seeds automatically and nondestructively to acquire the 3D model and phenotyping traits. A deeplabv3+ model with an Xception backbone was used to segment the sectional image of coconut fruits and seeds automatically. Compared with the structural-light system measurement, the mean absolute percentage error of the fruit volume and surface area measurements by the Micro-CT system was 1.87% and 2.24%, respectively, and the squares of the correlation coefficients were 0.977 and 0.964, respectively. In addition, compared with the manual measurements, the mean absolute percentage error of the automatic copra weight and total biomass measurements was 8.85% and 25.19%, respectively, and the adjusted squares of the correlation coefficients were 0.922 and 0.721, respectively. The Micro-CT system can nondestructively obtain up to 21 agronomic traits and 57 digital traits precisely.

Impacts of Urban Blue-Green Space on Residents' Health: A Bibliometric Review.

Urban blue-green space (UBGS), as an important component of the urban environment, is found to closely relate to human health. An extensive understanding of the effects of UBGS on human health is necessary for urban planning and intervention schemes towards healthy city development. However, a comprehensive review and discussion of relevant studies using bibliometric methods is still lacking. This paper adopted the bibliometric method and knowledge graph visualization technology to analyze the research on the impact of UBGS on residents' health, including the number of published papers, international influence, and network characteristics of keyword hotspots. The key findings include: (1) The number of articles published between 2001 and 2021 shows an increasing trend. Among the articles collected from WoS and CNKI, 38.74% and 32.65% of the articles focus on physical health, 38.32% and 30.61% on mental health, and 17.06% and 30.61% on public health, respectively. (2) From the analysis of international partnerships, countries with high levels of economic development and urbanization have closer cooperation than other countries. (3) UBGS has proven positive effects on residents' physical, mental, and public health. However, the mediating effects of UBGS on health and the differences in the health effects of UBGS on different ages and social classes are less studied. Therefore, this study proposes several future research directions. First, the mediating effect of UBGS on health impacts should be further examined. Furthermore, the interactive effects of residents' behaviors and the UBGS environment should be emphasized. Moreover, multidisciplinary integration should be strengthened. The coupling mechanism between human behavior and the environment should also be studied in depth with the help of social perception big data, wearable devices, and human-computer interactive simulation. Finally, this study calls for developing health risk monitoring and early warning systems, and integrating health impact assessment into urban planning, so as to improve residents' health and urban sustainability.

Video Crowd Localization With Multifocus Gaussian Neighborhood Attention and a Large-Scale Benchmark.

Video crowd localization is a crucial yet challenging task, which aims to estimate exact locations of human heads in the given crowded videos. To model spatial-temporal dependencies of human mobility, we propose a multi-focus Gaussian neighborhood attention (GNA), which can effectively exploit long-range correspondences while maintaining the spatial topological structure of the input videos. In particular, our GNA can also capture the scale variation of human heads well using the equipped multi-focus mechanism. Based on the multi-focus GNA, we develop a unified neural network called GNANet to accurately locate head centers in video clips by fully aggregating spatial-temporal information via a scene modeling module and a context cross-attention module. Moreover, to facilitate future researches in this field, we introduce a large-scale crowd video benchmark named VSCrowd (https://github.com/HopLee6/VSCrowd), which consists of 60K+ frames captured in various surveillance scenes and 2M+ head annotations. Finally, we conduct extensive experiments on three datasets including our VSCrowd, and the experiment results show that the proposed method is capable to achieve state-of-the-art performance for both video crowd localization and counting.

Effects of length and type of the alkyl chain on the micellization behavior of mixed systems of HS15 with fatty acids.

Due to volatility, low solubility and instability, the application of SCFAs and MCFAs is limited, which is expected to be solved by micelles. Taking SCFAs and MCFAs as models, this paper aims to research the influences of alkyl chain length and type on HS15 micelles. The critical micelle concentration (CMC) of various acid-HS15 systems was determined firstly. Then some air-water interface parameters and thermodynamic parameters were analyzed. Subsequently, particle size, cloud points (CP) and fluorophore release curves were measured. With the increase of CMC and Gmin, the decrease of size and CP, and the rapid quenching of fluorophores, it is more difficult for acids with longer-chain to form stable micelles with HS15 because of the strength of iceberg structure. Showing smaller CMC and Gmin, smaller size, higher CP and slower release of fluorescers, branched molecules can bind more closely to the hydrophobic part of HS15 due to their spatial flexibility.

Drug Resistance Mechanisms of Acute Myeloid Leukemia Stem Cells.

Acute myeloid leukemia (AML) is a polyclonal and heterogeneous hematological malignancy. Relapse and refractory after induction chemotherapy are still challenges for curing AML. Leukemia stem cells (LSCs), accepted to originate from hematopoietic stem/precursor cells, are the main root of leukemogenesis and drug resistance. LSCs are dynamic derivations and possess various elusive resistance mechanisms. In this review, we summarized different primary resistance and remolding mechanisms of LSCs after chemotherapy, as well as the indispensable role of the bone marrow microenvironment on LSCs resistance. Through a detailed and comprehensive review of the spectacle of LSCs resistance, it can provide better strategies for future researches on eradicating LSCs and clinical treatment of AML.

Correlation Analysis between Urban Elements and COVID-19 Transmission Using Social Media Data.

The outbreak of the COVID-19 has become a worldwide public health challenge for contemporary cities during the background of globalization and planetary urbanization. However, spatial factors affecting the transmission of the disease in urban spaces remain unclear. Based on geotagged COVID-19 cases from social media data in the early stage of the pandemic, this study explored the correlation between different infectious outcomes of COVID-19 transmission and various factors of the urban environment in the main urban area of Wuhan, utilizing the multiple regression model. The result shows that most spatial factors were strongly correlated to case aggregation areas of COVID-19 in terms of population density, human mobility and environmental quality, which provides urban planners and administrators valuable insights for building healthy and safe cities in an uncertain future.

TCGL: Temporal Contrastive Graph for Self-Supervised Video Representation Learning.

Video self-supervised learning is a challenging task, which requires significant expressive power from the model to leverage rich spatial-temporal knowledge and generate effective supervisory signals from large amounts of unlabeled videos. However, existing methods fail to increase the temporal diversity of unlabeled videos and ignore elaborately modeling multi-scale temporal dependencies in an explicit way. To overcome these limitations, we take advantage of the multi-scale temporal dependencies within videos and propose a novel video self-supervised learning framework named Temporal Contrastive Graph Learning (TCGL), which jointly models the inter-snippet and intra-snippet temporal dependencies for temporal representation learning with a hybrid graph contrastive learning strategy. Specifically, a Spatial-Temporal Knowledge Discovering (STKD) module is first introduced to extract motion-enhanced spatial-temporal representations from videos based on the frequency domain analysis of discrete cosine transform. To explicitly model multi-scale temporal dependencies of unlabeled videos, our TCGL integrates the prior knowledge about the frame and snippet orders into graph structures, i.e., the intra-/inter-snippet Temporal Contrastive Graphs (TCG). Then, specific contrastive learning modules are designed to maximize the agreement between nodes in different graph views. To generate supervisory signals for unlabeled videos, we introduce an Adaptive Snippet Order Prediction (ASOP) module which leverages the relational knowledge among video snippets to learn the global context representation and recalibrate the channel-wise features adaptively. Experimental results demonstrate the superiority of our TCGL over the state-of-the-art methods on large-scale action recognition and video retrieval benchmarks. The code is publicly available at https://github.com/YangLiu9208/TCGL.