<section class="abstract">
Spatial structures of proteins are closely related to protein functions. Integrating protein structures improves the performance of protein&ndash;protein interaction (PPI) prediction. However, the limited quantity of known protein structures restricts the application of structure-based prediction methods. Utilizing the predicted protein structure information is a promising method to improve the performance of sequence-based prediction methods. We propose a novel end-to-end framework, TAGPPI, to predict PPIs using protein sequence alone. TAGPPI extracts multi-dimensional features by employing 1D convolution operation on protein sequences and graph learning method on contact maps constructed from AlphaFold. A contact map contains abundant spatial structure information, which is difficult to obtain from 1D sequence data directly. We further demonstrate that the spatial information learned from contact maps improves the ability of TAGPPI in PPI prediction tasks. We compare the performance of TAGPPI with those of nine state-of-the-art sequence-based methods, and TAGPPI outperforms such methods in all metrics. To the best of our knowledge, this is the first method to use the predicted protein topology structure graph for sequence-based PPI prediction. More importantly, our proposed architecture could be extended to other prediction tasks related to proteins.
</section>
<div class="article-metadata-panel clearfix at-ArticleMetadata">&nbsp;</div>

Learning spatial structures of proteins improves protein–protein interaction prediction

A reduced removal of dysfunctional mitochondria is common to aging and age-related neurodegenerative pathologies such as Alzheimer&rsquo;s disease (AD). Strategies for treating such impaired mitophagy would benefit from the identification of mitophagy modulators. Here we report the combined use of unsupervised machine learning (involving vector representations of molecular structures, pharmacophore fingerprinting and conformer fingerprinting) and a cross-species approach for the screening and experimental validation of new mitophagy-inducing compounds. From a library of naturally occurring compounds, the workflow allowed us to identify 18 small molecules, and among them two potent mitophagy inducers (Kaempferol and Rhapontigenin). In nematode and rodent models of AD, we show that both mitophagy inducers increased the survival and functionality of glutamatergic and cholinergic neurons, abrogated amyloid-&beta; and tau pathologies, and improved the animals&rsquo; memory. Our findings suggest the existence of a conserved mechanism of memory loss across the AD models, this mechanism being mediated by defective mitophagy. The computational&ndash;experimental screening and validation workflow might help uncover potent mitophagy modulators that stimulate neuronal health and brain homeostasis.&nbsp;

Amelioration of Alzheimer’s disease pathology by mitophagy inducers identified via machine learning and a cross-species workflow

Using existing knowledge to carry out drug-disease associations prediction is a vital method for drug repositioning. However, effectively fusing the biomedical text and biological network information is one of the great challenges for most current drug repositioning methods. In this study, we propose a drug repositioning method based on heterogeneous networks and text mining (HeTDR). This model can combine drug features from multiple drug-related networks, disease features from biomedical corpora with the known drug-disease associations network to predict the correlation scores between drug and disease. Experiments demonstrate that HeTDR has excellent performance that is superior to that of state-of-the-art models. We present the top 10 novel HeTDR-predicted approved drugs for five diseases and prove our model is capable of discovering potential candidate drugs for disease indications.
&nbsp;

HeTDR: Drug repositioning based on heterogeneous networks and text mining

Geometry-based properties and characteristics of drug molecules play an important role in drug development for virtual screening in computational chemistry. The 3D characteristics of molecules largely determine the properties of the drug and the binding characteristics of the target. However, most of the previous studies focused on 1D or 2D molecular descriptors while ignoring the 3D topological structure, thereby degrading the performance of molecule-related prediction. Because it is very time-consuming to use dynamics to simulate molecular 3D conformer, we aim to use machine learning to represent 3D molecules by using the generated 3D molecular coordinates from the 2D structure.
We proposed Drug3D-Net, a novel deep neural network architecture based on the spatial geometric structure of molecules for predicting molecular properties. It is grid-based 3D convolutional neural network with spatial-temporal gated attention module, which can extract the geometric features for molecular prediction tasks in the process of convolution. The effectiveness of Drug3D-Net is verified on the public molecular datasets. Compared with other deep learning methods, Drug3D-Net shows superior performance in predicting molecular properties and biochemical activities.

A spatial-temporal gated attention module for molecular property prediction based on molecular geometry

<div id="abstract" class="content-section" data-v-2f5357c2="">
<div class="section-content" data-v-2f5357c2="">
<div data-v-2f5357c2="">
<div id="article--abstract--content" class="abstract">Finding target molecules with specific chemical properties plays a decisive role in drug development. We proposed GEOM-CVAE, a constrained variational autoencoder based on geometric representation for molecular generation with specific properties, which is protein-context-dependent. In terms of machine learning, it includes continuous feature embedding encoder and molecular generation decoder. Our key contribution is to propose an efficient geometric embedding method, including the spatial structure representations of drug molecule (converting the 3-D coordinates into image) and the geometric graph representations of protein target (modeling the protein surface as a mesh). The 3-D geometric information is vital to successful molecular generation, which is different from previous molecular generative methods based on 1-D or 2-D. Our model framework generates specific molecules in two phases, by first generating special image with molecular 3-D information to learn latent representations and generating molecules with constrained condition based on geometric graph convolution for specific protein and then inputting the generated structural molecules into a parser network for obtaining Simplified Molecular Input Line Entry System (SMILES) strings. Our model achieves competitive performance that implies its potential effectiveness to enable the exploration of the vast chemical space for drug discovery.</div>
</div>
</div>
</div>
<div id="full-text-links" class="content-section" data-v-2f5357c2=""></div>

Geometry-Based Molecular Generation With Deep Constrained Variational Autoencoder

Biomedical knowledge graphs (KGs), which can help with the understanding of complex biological systems and pathologies, have begun to play a critical role in medical practice and research. However, challenges remain in their embedding and use due to their complex nature and the specific demands of their construction. Existing studies often suffer from problems such as sparse and noisy datasets, insufficient modeling methods and non-uniform evaluation metrics. In this work, we established a comprehensive KG system for the biomedical field in an attempt to bridge the gap. Here, we introduced PharmKG, a multi-relational, attributed biomedical KG, composed of more than 500&nbsp;000 individual interconnections between genes, drugs and diseases, with 29 relation types over a vocabulary of ~8000 disambiguated entities. Each entity in PharmKG is attached with heterogeneous, domain-specific information obtained from multi-omics data, i.e. gene expression, chemical structure and disease word embedding, while preserving the semantic and biomedical features. For baselines, we offered nine state-of-the-art KG embedding (KGE) approaches and a new biological, intuitive, graph neural network-based KGE method that uses a combination of both global network structure and heterogeneous domain features. Based on the proposed benchmark, we conducted extensive experiments to assess these KGE models using multiple evaluation metrics. Finally, we discussed our observations across various downstream biological tasks and provide insights and guidelines for how to use a KG in biomedicine. We hope that the unprecedented quality and diversity of PharmKG will lead to advances in biomedical KG construction, embedding and application.

PharmKG: a dedicated knowledge graph benchmark for bomedical data mining

<div id="abstract-1" class="section abstract">
The drug repurposing of known approved drugs (e.g., lopinavir/ritonavir) has failed to treat SARS-CoV-2-infected patients. Therefore, it is important to generate new chemical entities against this virus. As a critical enzyme in the lifecycle of the coronavirus, the 3C-like main protease (3CLpro or Mpro) is the most attractive target for antiviral drug design. Based on a recently solved structure (PDB ID: 6LU7), we developed a novel advanced deep Q-learning network with a fragment-based drug design (ADQN&ndash;FBDD) for generating potential lead compounds targeting SARS-CoV-2 3CLpro. We obtained a series of derivatives from the lead compounds based on our structure-based optimization policy (SBOP). All of the 47 lead compounds obtained directly with our AI model and related derivatives based on the SBOP are accessible in our molecular library. These compounds can be used as potential candidates by researchers to develop drugs against SARS-CoV-2.
</div>

AI-aided design of novel targeted covalent inhibitors against SARS-CoV-2

The latest biological findings discover that the motionless 'lock-and-key' theory is no longer applicable and the flexibility of both the receptor and ligand plays a significant role in helping understand the principles of the binding affinity prediction. Based on this mechanism, molecular dynamics (MD) simulations have been invented as a useful tool to investigate the dynamical properties of this molecular system. However, the computational expenditure prohibits the growth of reported protein trajectories. To address this insufficiency, we present a novel spatial-temporal pre-training protocol, PretrainMD, to grant the protein encoder the capacity to capture the time-dependent geometric mobility along MD trajectories. Specifically, we introduce two sorts of self-supervised learning tasks: an atom-level denoising generative task and a protein-level snapshot ordering task. We validate the effectiveness of PretrainMD through the PDBbind dataset for both linear-probing and fine-tuning. Extensive experiments show that our PretrainMD exceeds most state-of-the-art methods and achieves comparable performance. More importantly, through visualization we discover that the learned representations by pre-training on MD trajectories without any label from the downstream task follow similar patterns of the magnitude of binding affinities. This strongly aligns with the fact that the motion of the interactions of protein and ligand maintains the key information of their binding. Our work provides a promising perspective of self-supervised pre-training for protein representations with very fine temporal resolutions and hopes to shed light on the further usage of MD simulations for the biomedical deep learning community.
<div class="metatable">&nbsp;</div>

Pre-training of Deep Protein Models with Molecular Dynamics Simulations for Drug Binding

Recent literature suggested that&nbsp;ALDH2&nbsp;mutation is associated with alcohol metabolism, and ethanol intake might jointly increase the risk of Alzheimer&rsquo;s disease (AD) in mice. However, it is unclear whether this synergistic effect exists among humans. We examined the associations of four single nucleotide polymorphisms (SNPs) on aldehyde dehydrogenase (ALDH) and alcohol dehydrogenase (ADH) genes (i.e.,&nbsp;ALDH2&nbsp;rs671,&nbsp;ADH1B&nbsp;rs1229984,&nbsp;ADH1B&nbsp;rs1042026, and&nbsp;ADH1C&nbsp;rs1693482) and cognitive impairment among the oldest-old. We also investigated whether this association was modified by ethanol intake from alcohol consumption. Data were from the Chinese Longitudinal Healthy Longevity Survey genetic sub-study, including 1,949 participants aged over 90 years. Participants with a Mini-Mental State Examination (MMSE) score of &lt; 18 were considered cognitively impaired. Alcohol consumption was categorized as heavy, moderate, or never drinkers. With the dominant model, carrying A allele on rs671, C allele on rs1229984, and T allele on rs1042026 was associated with 33% (95% confidence interval [CI]: 5%, 69%), 33% (95% CI: 2%, 75%), and 29% (95% CI: 3%, 62%) higher odds of cognitive impairment in the multivariable-adjusted logistic model, respectively. We did not observe a significant interaction between those SNPs and alcohol consumption. Among the oldest-old, carrying&nbsp;ALDH2&nbsp;rs671 mutation was associated with higher odds of cognitive impairment independent of alcohol consumption.

Associations of Alcohol Dehydrogenase and Aldehyde Dehydrogenase Polymorphism with Cognitive Impairment among the Oldest Old in China

Book chapter: Aging and Alzheimer's Disease
Artificial intelligence (AI) implies the use of a machine with limited human interference to model intelligent actions. It covers a broad range of research studies from machine intelligence for computer vision, robotics, and natural language processing to more theoretical machine learning algorithm design and, recently, &ldquo;deep learning&rdquo; development. The application of AI in medical fields is booming, including the use of AI in data collection, analysis, mechanistic prediction, to clinical disease diagnosis and drug development. In this chapter, we focus on the challenges in the studies of aging and the age-predisposed Alzheimer&rsquo;s disease (AD) and summarize on how to use AI to help addressing these questions. We finally provide future perspectives on the use of AI in aging research and AD.

Artificial Intelligence in Medicine - Chapter: Ageing and Alzheimer’s Disease

The rapid spread of coronavirus 2019 disease (COVID-19) has manifested a global public health crisis, and chest CT has been proven to be a powerful tool for screening, triage, evaluation and prognosis in COVID-19 patients. However, CT is not only costly but also associated with an increased incidence of cancer, in particular for children. This study will question whether clinical symptoms and laboratory results can predict the CT outcomes for the pediatric patients with positive RT-PCR testing results in order to determine the necessity of CT for such a vulnerable group. Clinical data were collected from 244 consecutive pediatric patients (16 years of age and under) treated at Wuhan Children's Hospital with positive RT-PCR testing, and the chest CT were performed within 3 days of clinical data collection, from January 21 to March 8, 2020. This study was approved by the local ethics committee of Wuhan Children's Hospital. Advanced decision tree based machine learning models were developed for the prediction of CT outcomes. Results have shown that age, lymphocyte, neutrophils, ferritin and C-reactive protein are the most related clinical indicators for predicting CT outcomes for pediatric patients with positive RT-PCR testing. Our decision support system has managed to achieve an AUC of 0.84 with 0.82 accuracy and 0.84 sensitivity for predicting CT outcomes. Our model can effectively predict CT outcomes, and our findings have indicated that the use of CT should be reconsidered for pediatric patients, as it may not be indispensable.

Can clinical symptoms and laboratory results predict CT abnormality? initial findings using novel machine learning techniques in children with COVID-19 infections

<section lang="en" aria-labelledby="Abs1" data-title="Abstract" data-gtm-vis-first-on-screen-50443292_562="422" data-gtm-vis-total-visible-time-50443292_562="5700" data-gtm-vis-first-on-screen-50443292_563="422" data-gtm-vis-total-visible-time-50443292_563="5700" data-gtm-vis-polling-id-50443292_563="2089" data-gtm-vis-polling-id-50443292_562="2090" data-gtm-vis-recent-on-screen-50443292_563="18814" data-gtm-vis-recent-on-screen-50443292_562="18814">
<div id="Abs1-section" class="c-article-section">
<div id="Abs1-content" class="c-article-section__content">
Machine learning methods offer great promise for fast and accurate detection and prognostication of coronavirus disease 2019 (COVID-19) from standard-of-care chest radiographs (CXR) and chest computed tomography (CT) images. Many articles have been published in 2020 describing new machine learning-based models for both of these tasks, but it is unclear which are of potential clinical utility. In this systematic review, we consider all published papers and preprints, for the period from 1 January 2020 to 3 October 2020, which describe new machine learning models for the diagnosis or prognosis of COVID-19 from CXR or CT images. All manuscripts uploaded to bioRxiv, medRxiv and arXiv along with all entries in EMBASE and MEDLINE in this timeframe are considered. Our search identified 2,212 studies, of which 415 were included after initial screening and, after quality screening, 62 studies were included in this systematic review. Our review finds that none of the models identified are of potential clinical use due to methodological flaws and/or underlying biases. This is a major weakness, given the urgency with which validated COVID-19 models are needed. To address this, we give many recommendations which, if followed, will solve these issues and lead to higher-quality model development and well-documented manuscripts.
</div>
</div>
</section>
<section data-title="Main" data-gtm-vis-first-on-screen-50443292_563="2490" data-gtm-vis-total-visible-time-50443292_563="2900" data-gtm-vis-first-on-screen-50443292_562="2490" data-gtm-vis-total-visible-time-50443292_562="2900" data-gtm-vis-polling-id-50443292_563="2285" data-gtm-vis-polling-id-50443292_562="2286" data-gtm-vis-recent-on-screen-50443292_563="19680" data-gtm-vis-recent-on-screen-50443292_562="19680">
<div id="Sec1-section" class="c-article-section"></div>
</section>

Common pitfalls and recommendations for using machine learning to detect and prognosticate for COVID-19 using chest radiographs and CT scans

A drug-like-molecule library can contain 1023&ndash;1060&nbsp;molecules, among which only approximately 1012&nbsp;molecules may be synthesized in labs. However, it is still challenging for researchers to find the most promising candidates among the vast number of synthesizable compounds in a reasonable time. Moreover, although molecules are picked for their predicted bioactivities, their absorption, distribution, metabolism, excretion, and toxicity (ADMET) properties are often difficult to predict and modify. This is often a bottleneck for downstream studies and applications. It would be more productive if candidate molecules are generated, rather than screened from libraries, with suitable ADMET properties as prerequisites at the beginning of the molecule design process. Recently, artificial intelligence (AI)-based generative models have been described for designing drug candidates using prior biological and chemical knowledge. A spectacular example was the use of a combination of AI generative techniques and reinforcement learning by the biotechnology company, Insilico Medicine, to successfully create new DDR1 kinase inhibitors to treat fibrosis in only 21&nbsp;days. We will describe how reinforcement learning (RL) algorithms can be applied to generative AI for better real-world effectiveness while better utilizing modern distributed hardware assets. In this chapter, we will review simple and advanced AI generative models and discuss the advantages and disadvantages of each model.

Generative AI Models for Drug Discovery

Background:&nbsp;Apolipoprotein E (APOE) &epsilon;4 is the single most important genetic risk factor for cognitive impairment and Alzheimer disease (AD), while lifestyle factors such as smoking, drinking, diet, and physical activity also have impact on cognition. The goal of the study is to investigate whether the association between lifestyle and cognition varies by APOE genotype among the oldest old.
Methods and findings:&nbsp;We used the cross-sectional data including 6,160 oldest old (aged 80 years old or older) from the genetic substudy of the Chinese Longitudinal Healthy Longevity Survey (CLHLS) which is a national wide cohort study that began in 1998 with follow-up surveys every 2-3 years. Cognitive impairment was defined as a Mini-Mental State Examination (MMSE) score less than 18. Healthy lifestyle profile was classified into 3 groups by a composite measure including smoking, alcohol consumption, dietary pattern, physical activity, and body weight. APOE genotype was categorized as APOE &epsilon;4 carriers versus noncarriers. We examined the associations of cognitive impairment with lifestyle profile and APOE genotype using multivariable logistic regressions, controlling for age, sex, education, marital status, residence, disability, and numbers of chronic conditions. The mean age of our study sample was 90.1 (standard deviation [SD], 7.2) years (range 80-113); 57.6% were women, and 17.5% were APOE &epsilon;4 carriers. The mean MMSE score was 21.4 (SD: 9.2), and 25.0% had cognitive impairment. Compared with those with an unhealthy lifestyle, participants with intermediate and healthy lifestyle profiles were associated with 28% (95% confidence interval [CI]: 16%-38%, P &lt; 0.001) and 55% (95% CI: 44%-64%, P &lt; 0.001) lower adjusted odds of cognitive impairment. Carrying the APOE &epsilon;4 allele was associated with 17% higher odds (95% CI: 1%-31%, P = 0.042) of being cognitively impaired in the adjusted model. The association between lifestyle profiles and cognitive function did not vary significantly by APOE &epsilon;4 genotype (noncarriers: 0.47 [0.37-0.60] healthy versus unhealthy; carriers: 0.33 [0.18-0.58], P for interaction = 0.30). The main limitation was the lifestyle measurements were self-reported and were nonspecific. Generalizability of the findings is another limitation because the study sample was from the oldest old in China, with unique characteristics such as low body weight compared to populations in high-income countries.
Conclusions:&nbsp;In this study, we observed that healthier lifestyle was associated with better cognitive function among the oldest old regardless of APOE genotype. Our findings may inform the cognitive outlook for those oldest old with high genetic risk of cognitive impairment.

Association of APOE ε4 genotype and lifestyle with cognitive function among Chinese adults aged 80 years and older: A cross-sectional study

Studies have shown contact with nature has positive psychological, neurological, and cognitive benefits. Whether the built environment can affect genetic predisposition of Alzheimer's disease (AD) should be explored. We aimed to examine whether greenness around the residential environment can modify the effect of genetic AD risk on cognitive function. We used a genetic sub-study of the Chinese Longitudinal Healthy Longevity Survey including 1199 older adults (mean age: 100.3 &plusmn; 3.4 years) aged 90 years old or older. We used Polygenic Risk Score (PRS) to quantify the genetic AD risk and two types of measurements based on Normalized Difference Vegetation Index (NDVI) to access the residential greenness (contemporaneous and annual average NDVI). Contemporaneous NDVI values were the NDVI value collected at the corresponding survey, and the annual average NDVI was the average value of NDVI during the year before the corresponding survey. We defined cognitive impairment as having a Mini-Mental State Examination score below 25. In the multivariable logistics regression models, contemporaneous NDVI and genetic AD risk were associated with cognitive impairment. Among those with low genetic AD risk, the risk of cognitive impairment was lower in those living around higher greenness (contemporaneous NDVI OR: 0.55, 95% CI: [0.34, 0.86]; Pinteraction: 0.071; annual average NDVI OR: 0.49, 95% CI: [0.31, 0.79]; Pinteraction: 0.040). We did not observe significant associations between greenness and cognitive impairment among those with high genetic AD risk. Prevention efforts using PRS warrant a higher granularity of environmental exposures and biological etiology data.

Interaction of greenness and polygenic risk score of Alzheimer's disease on risk of cognitive impairment

Introduction:&nbsp;The interactions between apolipoprotein E (APOE) genotype and diet pattern changes were found significant in several trials, implying that&nbsp;APOE&nbsp;gene may modify the effect of animal protein-rich food on health outcomes. We aim to study the interaction of&nbsp;APOE&nbsp;genotype with the effect of meat, fish and egg intake on mortality.
Methods:&nbsp;This population-based study enrolled 8,506 older adults (mean age: 81.7 years, 52.3% female) from the Chinese Longitudinal Healthy Longevity Study. The intake frequency of meat, fish and egg was assessed by 3-point questions at baseline. Cox regression was conducted to calculate the hazard ratios for all-cause mortality of intake levels of meat, fish and egg. The analyses were stratified by&nbsp;APOE&nbsp;genotype and sex. The analyses were performed in 2020.
Results:&nbsp;In the multivariable-adjusted models, meat and fish intake was associated with all-cause mortality (high vs. low intake: meat: HR: 1.14, 95% CI: 1.01, 1.28; fish: HR: 0.83, 95% CI: 0.73, 0.95).&nbsp;APOE&nbsp;genotype have significant interactions with meat and fish intake (Ps &lt; 0.05). Compared with low fish intake, high fish intake was associated with lower risk of mortality (HR: 0.74, 95% CI: 0.56&ndash;0.98) only among the&nbsp;APOE&nbsp;&epsilon;4 carriers. High meat intake was significantly associated with higher risks of mortality (HR: 1.13, 95% CI: 1.04&ndash;1.25) only among the&nbsp;APOE&nbsp;&epsilon;4 non-carriers. The interactive relationship was restricted among the male. No significant findings were observed between egg and mortality among carriers or non-carriers.
Conclusions:&nbsp;Among Chinese older adults, the significance of associations of mortality with reported meat or fish intake depended on APOE-E4 carriage status. If validated by other studies, our findings provide evidence for gene-based &ldquo;precision&rdquo; lifestyle recommendations.

Apolipoprotein E Genotype, Meat, Fish, and Egg Intake in Relation to Mortality Among Older Adults: A Longitudinal Analysis in China

Objective: Biological age (BA) has been accepted as a more accurate proxy of aging than chronological age (CA). This study aimed to use machine learning (ML) algorithms to estimate BA in the Chinese population. Materials and Methods: We used data from 9,771 middle-aged and older (&ge; 45 years) Chinese adults (&ge; 45 years) in the 2011/2012 wave of the China Health and Retirement Longitudinal Study and followed until 2018. We used several ML algorithms (e.g., Gradient Boosting Regressor, Random Forest, CatBoost Regressor, and Support Vector Machine) to develop new measures of biological aging (ML-BAs) based on physiological biomarkers. R-squared value and mean absolute error (MAE) were used to determine the optimal performance of these ML-BAs. We used logistic regression models to examine the associations of the best ML-BA and a conventional aging measure &ndash; Klemera and Doubal method-biological age (KDM-BA) we previously developed &ndash; with physical disability and mortality, respectively. Results: The Gradient Boosting Regression model performed the best, resulting in an ML-BA with an R-squared value of 0.270 and MAE of 6.519. This ML-BA was significantly associated with disability in basic activities of daily living, instrumental activities of daily living, lower extremity mobility, and upper extremity mobility, and mortality, with odds ratios ranging from 1% to 7% (per one-year increment in ML-BA, all P &lt;0.001), independent of CA. These associations were generally comparable to that of KDM-BA. Conclusion: This study provides a valid ML-based measure of biological aging for middle-aged and older Chinese adults. These findings support the application of ML in geroscience research and may help facilitate preventive and geroprotector intervention studies.

Machine learning-based aging measure among middle-aged and older Chinese adults: the China Health and Retirement Longitudinal Study

One of the key issues facing public healthcare is the global trend of an increasingly ageing society which continues to present policy makers and caregivers with formidable healthcare and socio-economic challenges. Ageing is the primary contributor to a broad spectrum of chronic disorders all associated with a lower quality of life in the elderly. In 2019, the Chinese population constituted 18 % of the world population, with 164.5 million Chinese citizens aged 65 and above (65+), and 26 million aged 80 or above (80+). China has become an ageing society, and as it continues to age it will continue to exacerbate the burden borne by current family and public healthcare systems. Major healthcare challenges involved with caring for the elderly in China include the management of chronic non-communicable diseases (CNCDs), physical frailty, neurodegenerative diseases, cardiovascular diseases, with emerging challenges such as providing sufficient dental care, combating the rising prevalence of sexually transmitted diseases among nursing home communities, providing support for increased incidences of immune diseases, and the growing necessity to provide palliative care for the elderly. At the governmental level, it is necessary to make long-term strategic plans to respond to the pressures of an ageing society, especially to establish a nationwide, affordable, annual health check system to facilitate early diagnosis and provide access to affordable treatments. China has begun work on several activities to address these issues including the recent completion of the of the Ten-year Health-Care Reform project, the implementation of the Healthy China 2030 Action Plan, and the opening of the National Clinical Research Center for Geriatric Disorders. There are also societal challenges, namely the shift from an extended family system in which the younger provide home care for their elderly family members, to the current trend in which young people are increasingly migrating towards major cities for work, increasing reliance on nursing homes to compensate, especially following the outcomes of the &lsquo;one child policy&rsquo; and the &lsquo;empty-nest elderly&rsquo; phenomenon. At the individual level, it is important to provide avenues for people to seek and improve their own knowledge of health and disease, to encourage them to seek medical check-ups to prevent/manage illness, and to find ways to promote modifiable health-related behaviors (social activity, exercise, healthy diets, reasonable diet supplements) to enable healthier, happier, longer, and more productive lives in the elderly. Finally, at the technological or treatment level, there is a focus on modern technologies to counteract the negative effects of ageing. Researchers are striving to produce drugs that can mimic the effects of &lsquo;exercising more, eating less&rsquo;, while other anti-ageing molecules from molecular gerontologists could help to improve &lsquo;healthspan&rsquo; in the elderly. Machine learning, &lsquo;Big Data&rsquo;, and other novel technologies can also be used to monitor disease patterns at the population level and may be used to inform policy design in the future. Collectively, synergies across disciplines on policies, geriatric care, drug development, personal awareness, the use of big data, machine learning and personalized medicine will transform China into a country that enables the most for its elderly, maximizing and celebrating their longevity in the coming decades. This is the 2nd edition of the review paper.

A research agenda for ageing in China in the 21st century: focusing on basic and translational research, long-term care, policy and social networks

Background:&nbsp;Cardiovascular disease is the leading cause of death among women worldwide, yet, women have historically been underrepresented in cardiovascular trials.
&nbsp;
Methods:&nbsp;We systematically assessed the participation of women in completed cardiovascular trials registered in ClinicalTrials.gov between 2010 and 2017, and extracted publicly available information including disease type, sponsor type, country, trial size, intervention type, and the demographic characteristics of trial participants. We calculated the female-to-male ratio for each trial and determined the prevalence-adjusted estimates for participation of women by dividing the percentage of women among trial participants by the percentage of women in the disease population (participation prevalence ratio; a ratio of 0.8 to 1.2 suggests comparable prevalence and good representation).
&nbsp;
Results:&nbsp;We identified 740 completed cardiovascular trials including a total of 862 652 adults, of whom 38.2% were women. The median female-to-male ratio of each trial was 0.51 (25th quartile, 0.32; 75th quartile, 0.90) overall and varied by age group (1.02 in &le;55 year old group versus 0.40 in the 61- to 65-year-old group), type of intervention (0.44 for procedural trials versus 0.78 for lifestyle intervention trials), disease type (0.34 for acute coronary syndrome versus 3.20 for pulmonary hypertension), region (0.45 for Western Pacific versus 0.55 for the Americas), funding/sponsor type (0.14 for government-funded versus 0.73 for multiple sponsors), and trial size (0.56 for smaller [n&le;47] versus 0.49 for larger [n&ge;399] trials). Relative to their prevalence in the disease population, participation prevalence ratio was higher than 0.8 for hypertension, pulmonary arterial hypertension and lower (participation prevalence ratio 0.48 to 0.78) for arrhythmia, coronary heart disease, acute coronary syndrome, and heart failure trials. The most recent time period (2013 to 2017) saw significant increases in participation prevalence ratios for stroke (P=0.007) and heart failure (P=0.01) trials compared with previous periods.
&nbsp;
Conclusions:&nbsp;Among cardiovascular trials in the current decade, men still predominate overall, but the representation of women varies with disease and trial characteristics, and has improved in stroke and heart failure trials.

Women's participation in cardiovascular clinical trials from 2010 to 2017

Constructing proper representations of molecules lies at the core of numerous tasks such as molecular property prediction and drug design. Graph neural networks, especially message passing neural network (MPNN) and its variants, have recently made remarkable achievements in molecular graph modeling. Albeit powerful, the one-sided focuses on atom (node) or bond (edge) information of existing MPNN methods lead to the insufficient representations of the attributed molecular graphs. Herein, we propose a Communicative Message Passing Neural Network (CMPNN) to improve the molecular embedding by strengthening the message interactions between nodes and edges through a communicative kernel. In addition, the message generation process is enriched by introducing a new message booster module. Extensive experiments demonstrated that the proposed model obtained superior performances against state-of-the-art baselines on six chemical property datasets. Further visualization also showed better representation capacity of our model.

Communicative representation learning on attributed molecular graphs

Efficient and accurate prediction of molecular properties, such as lipophilicity and solubility, is highly desirable for rational compound design in chemical and pharmaceutical industries. To this end, we build and apply a graph-neural-network framework called self-attention-based message-passing neural network (SAMPN) to study the relationship between chemical properties and structures in an interpretable way. The main advantages of SAMPN are that it directly uses chemical graphs and breaks the black-box mold of many machine/deep learning methods. Specifically, its attention mechanism indicates the degree to which each atom of the molecule contributes to the property of interest, and these results are easily visualized. Further, SAMPN outperforms random forests and the deep learning framework MPN from Deepchem. In addition, another formulation of SAMPN (Multi-SAMPN) can simultaneously predict multiple chemical properties with higher accuracy and efficiency than other models that predict one specific chemical property. Moreover, SAMPN can generate chemically visible and interpretable results, which can help researchers discover new pharmaceuticals and materials. The source code of the SAMPN prediction pipeline is freely available at Github (<a href="https://github.com/tbwxmu/SAMPN">https://github.com/tbwxmu/SAMPN</a>).

A self-attention based message passing neural network for predicting molecular lipophilicity and aqueous solubility

Accurate prediction of molecular properties is important for new compound design, which is a crucial step in drug discovery. In this paper, molecular graph data is utilized for property prediction based on graph convolution neural networks. In addition, a convolution spatial graph embedding layer (C-SGEL) is introduced to retain the spatial connection information on molecules. And, multiple C-SGELs are stacked to construct a convolution spatial graph embedding network (C-SGEN) for end-to-end representation learning. In order to enhance the robustness of the network, molecular fingerprints are also combined with C-SGEN to build a composite model for predicting molecular properties. Our comparative experiments have shown that our method is accurate and achieves the best results on some open benchmark datasets.

论文

Featured Publications