Graph is an important data representation which appears in a wide diversity of real-world scenarios. Effective graph analytics provides users a deeper understanding of what is behind the data, and ...thus can benefit a lot of useful applications such as node classification, node recommendation, link prediction, etc. However, most graph analytics methods suffer the high computation and space cost. Graph embedding is an effective yet efficient way to solve the graph analytics problem. It converts the graph data into a low dimensional space in which the graph structural information and graph properties are maximumly preserved. In this survey, we conduct a comprehensive review of the literature in graph embedding. We first introduce the formal definition of graph embedding as well as the related concepts. After that, we propose two taxonomies of graph embedding which correspond to what challenges exist in different graph embedding problem settings and how the existing work addresses these challenges in their solutions. Finally, we summarize the applications that graph embedding enables and suggest four promising future research directions in terms of computation efficiency, problem settings, techniques, and application scenarios.
Transition metal dichalcogenides (TMDs) show wide ranges of electronic properties ranging from semiconducting, semi-metallic to metallic due to their remarkable structural differences. To obtain 2D ...TMDs with specific properties, it is extremely important to develop particular strategies to obtain specific phase structures. Phase engineering is a traditional method to achieve transformation from one phase to another controllably. Control of such transformations enables the control of properties and access to a range of properties, otherwise inaccessible. Then extraordinary structural, electronic and optical properties lead to a broad range of potential applications. In this review, we introduce the various electronic properties of 2D TMDs and their polymorphs, and strategies and mechanisms for phase transitions, and phase transition kinetics. Moreover, the potential applications of 2D TMDs in energy storage and conversion, including electro/photocatalysts, batteries/supercapacitors and electronic devices, are also discussed. Finally, opportunities and challenges are highlighted. This review may further promote the development of TMD phase engineering and shed light on other two-dimensional materials of fundamental interest and with potential ranges of applications.
The diversity of electronic characteristics of TMDs ranging from the semiconducting, semi-metallic to metallic have broadened their application in catalysis, electrode materials and next-generation functional electronic devices.
Display omitted
•A systematic review of the current works pertinent to patient representation learning.•A growing trend in building deep learning based patient representations from EHRs.•The learned ...representations attempt to gain a cohesive picture of a patient’s data.•Capabilities of deep learning models can largely address the challenges of EHR data.•Future work: advanced learning methods to obtain robust, and precise representations.
Patient representation learning refers to learning a dense mathematical representation of a patient that encodes meaningful information from Electronic Health Records (EHRs). This is generally performed using advanced deep learning methods. This study presents a systematic review of this field and provides both qualitative and quantitative analyses from a methodological perspective.
We identified studies developing patient representations from EHRs with deep learning methods from MEDLINE, EMBASE, Scopus, the Association for Computing Machinery (ACM) Digital Library, and the Institute of Electrical and Electronics Engineers (IEEE) Xplore Digital Library. After screening 363 articles, 49 papers were included for a comprehensive data collection.
Publications developing patient representations almost doubled each year from 2015 until 2019. We noticed a typical workflow starting with feeding raw data, applying deep learning models, and ending with clinical outcome predictions as evaluations of the learned representations. Specifically, learning representations from structured EHR data was dominant (37 out of 49 studies). Recurrent Neural Networks were widely applied as the deep learning architecture (Long short-term memory: 13 studies, Gated recurrent unit: 11 studies). Learning was mainly performed in a supervised manner (30 studies) optimized with cross-entropy loss. Disease prediction was the most common application and evaluation (31 studies). Benchmark datasets were mostly unavailable (28 studies) due to privacy concerns of EHR data, and code availability was assured in 20 studies.
The existing predictive models mainly focus on the prediction of single diseases, rather than considering the complex mechanisms of patients from a holistic review. We show the importance and feasibility of learning comprehensive representations of patient EHR data through a systematic review. Advances in patient representation learning techniques will be essential for powering patient-level EHR analyses. Future work will still be devoted to leveraging the richness and potential of available EHR data. Reproducibility and transparency of reported results will hopefully improve. Knowledge distillation and advanced learning techniques will be exploited to assist the capability of learning patient representation further.
Interleukin-17 (IL-17), IL-21, IL-22 and IL-23 can be grouped as T helper 17 (Th17)-related cytokines because they are either produced by Th17/Th22 cells or involved in their development. Here, we ...review Th17-related cytokines/Th17-like cells, networks/signals and their roles in immune responses or immunity against Mycobacterium tuberculosis (Mtb) infection. Published studies suggest that Th17-related cytokine pathways may be manipulated by Mtb microorganisms for their survival benefits in primary tuberculosis (TB). In addition, there is evidence that immune responses of the signal transducer and activator of transcription 3 (STAT3) signal pathway and Th17-like T-cell subsets are dysregulated or destroyed in patients with TB. Furthermore, Mtb infection can impact upstream cytokines in the STAT3 pathway of Th17-like responses. Based on these findings, we discuss the need for future studies and the rationale for targeting Th17-related cytokines/signals as a potential adjunctive treatment.
We analyzed the adsorption of Li on graphene in the context of anodes for lithium-ion batteries (LIBs) using first-principles methods including van der Waals interactions. We found that although Li ...can reside on the surface of defect-free graphene under favorable conditions, the binding is much weaker than to graphite and the concentration on a graphene surface is not higher than in graphite. At low concentration, Li ions spread out on graphene because of Coulomb repulsion. With increased Li content, we found that small Li clusters can be formed on graphene. Although this result suggests that graphene nanosheets can conceivably have a higher ultimate Li capacity than graphite, it should be noted that such nanoclusters can potentially nucleate Li dendrites, leading to failure. The implications for nanostructured carbon anodes in batteries are discussed.
We previously reported preliminary findings that post induction imatinib mesylate (340 mg/m(2)/day), in combination with intensive chemotherapy, resulted in outcomes similar to blood and marrow ...transplant (BMT) for pediatric patients with Philadelphia chromosome-positive (Ph+) acute lymphoblastic leukemia (ALL). We now report 5-year outcomes of imatinib plus intensive chemotherapy in 91 children (1-21 years) with and without allogeneic BMT (N=91). We explore the impacts of additional chromosomal abnormalities and minimal residual disease (MRD) by flow cytometry on outcomes. The 5-year disease-free survival was similar for Cohort 5 patients, treated with chemotherapy plus imatinib (70%±12%, n=28), sibling donor BMT patients (65%±11%, n=21) and unrelated donor BMT patients (59±15%; P=0.60, n=13). Patients with additional cytogenetic abnormalities had worse outcomes (P=0.05). End induction (pre-imatinib) MRD was not prognostic for Cohort 5 or allogeneic BMT patients, although limited by small numbers. The re-induction rate following relapse was similar to other higher-risk ALL groups. Longer-term follow-up confirms our initial observation of substantially good outcomes for children and adolescents with Ph+ ALL treated with imatinib plus intensive chemotherapy with no advantage for allogeneic BMT.
With first-principles DFT calculations, the interaction between Li and carbon in graphene-based nanostructures is investigated as Li is adsorbed on graphene. It is found that the Li/C ratio of less ...than 1/6 for the single-layer graphene is favorable energetically, which can explain what has been observed in Raman spectrum reported recently. In addition, it is also found that the pristine graphene cannot enhance the diffusion energetics of Li ion. However, the presence of vacancy defects can increase the ratio of Li/C largely. With double-vacancy and higher-order defects, Li ion can diffuse freely in the direction perpendicular to the graphene sheets and hence boost the diffusion energetics to some extent.
The article discusses the relationship between biomedical research, informatics, data science as well as artificial intelligence (AI). The most significant advancement is the approach driven by AI ...models, which is different from traditional hypothesis-based research. Some of the ways in which large amounts of medical data could be put to use are highlighted.
Tumor-derived DNA can be found in the plasma of cancer patients. In this study, we explored the use of shotgun massively parallel sequencing (MPS) of plasma DNA from cancer patients to scan a cancer ...genome noninvasively.
Four hepatocellular carcinoma patients and a patient with synchronous breast and ovarian cancers were recruited. DNA was extracted from the tumor tissues, and the preoperative and postoperative plasma samples of these patients were analyzed with shotgun MPS.
We achieved the genomewide profiling of copy number aberrations and point mutations in the plasma of the cancer patients. By detecting and quantifying the genomewide aggregated allelic loss and point mutations, we determined the fractional concentrations of tumor-derived DNA in plasma and correlated these values with tumor size and surgical treatment. We also demonstrated the potential utility of this approach for the analysis of complex oncologic scenarios by studying the patient with 2 synchronous cancers. Through the use of multiregional sequencing of tumoral tissues and shotgun sequencing of plasma DNA, we have shown that plasma DNA sequencing is a valuable approach for studying tumoral heterogeneity.
Shotgun DNA sequencing of plasma is a potentially powerful tool for cancer detection, monitoring, and research.
Although Hi-C technology is one of the most popular tools for studying 3D genome organization, due to sequencing cost, the resolution of most Hi-C datasets are coarse and cannot be used to link ...distal regulatory elements to their target genes. Here we develop HiCPlus, a computational approach based on deep convolutional neural network, to infer high-resolution Hi-C interaction matrices from low-resolution Hi-C data. We demonstrate that HiCPlus can impute interaction matrices highly similar to the original ones, while only using 1/16 of the original sequencing reads. We show that the models learned from one cell type can be applied to make predictions in other cell or tissue types. Our work not only provides a computational framework to enhance Hi-C data resolution but also reveals features underlying the formation of 3D chromatin interactions.
PDF
