Development of a molecular feature-based survival prediction model of ovarian cancer using the deep neural network

Ovarian cancer(OC)is one of the most lethal gynecologic cancer worldwide,and survival prediction is meaningful for personalized treatment.^(1)The survival outcome of cancer patients mainly depended on the malignancy of the primary tumor which is tightly linked with the expression profile of the molecular features.^(2)Therefore,in this study,we developed a molecular feature-based survival prediction model of OC using a deep neural network(DNN).

Efficient Preference Clustering via Random Fourier Features

Approximations based on random Fourier features have recently emerged as an efficient and elegant method for designing large-scale machine learning tasks.Unlike approaches using the Nystr?m method,which randomly samples the training examples,we make use of random Fourier features,whose basis functions(i.e.,cosine and sine)are sampled from a distribution independent from the training sample set,to cluster preference data which appears extensively in recommender systems.Firstly,we propose a two-stage preference clustering framework.In this framework,we make use of random Fourier features to map the preference matrix into the feature matrix,soon afterwards,utilize the traditional k-means approach to cluster preference data in the transformed feature space.Compared with traditional preference clustering,our method solves the problem of insufficient memory and greatly improves the efficiency of the operation.Experiments on movie data sets containing 100000 ratings,show that the proposed method is more effective in clustering accuracy than the Nystr?m and k-means,while also achieving better performance than these clustering approaches.

Establishment of an indicator framework for global One Health Intrinsic Drivers index based on the grounded theory and fuzzy analytical hierarchy-entropy weight method

Background:One Health has become a global consensus to deal with complex health problems.However,the pro‑gress of One Health implementation in many countries is still relatively slow,and there is a lack of systematic evalua‑tion index.The purpose of this study was to establish an indicator framework for global One Health Intrinsic Drivers index(GOH-IDI)to evaluate human,animal and environmental health development process globally.Method:First,82 studies were deeply analyzed by a grounded theory(GT)method,including open coding,axial coding,and selective coding,to establish a three-level indicator framework,which was composed of three selective codes,19 axial codes,and 79 open codes.Then,through semi-structured interviews with 28 health-related experts,the indicators were further integrated and simplifed according to the inclusion criteria of the indicators.Finally,the fuzzy analytical hierarchy process combined with the entropy weight method was used to assign weights to the indi‑cators,thus,forming the evaluation indicator framework of human,animal and environmental health development process.Results:An indicator framework for GOH-IDI was formed consisting of three selective codes,15 axial codes and 61 open codes.There were six axial codes for“Human Health”,of which“Infectious Diseases”had the highest weight(19.76%)and“Injuries and Violence”had the lowest weight(11.72%).There were four axial codes for“Animal Health”,of which“Animal Epidemic Disease”had the highest weight(39.28%)and“Animal Nutritional Status”had the low‑est weight(11.59%).Five axial codes were set under“Environmental Health”,among which,“Air Quality and Climate Change”had the highest weight(22.63%)and“Hazardous Chemicals”had the lowest weight(17.82%).Conclusions:An indicator framework for GOH-IDI was established in this study.The framework were universal,balanced,and scientifc,which hopefully to be a tool for evaluation of the joint development of human,animal and environmental health in diferent regions globally.

Disruption of KDM4C-ALDH1 A3 feed-forward loop inhibits sternness,tumorigenesis and chemoresistance of gastric cancer stem cells

Dear Editor,Gastric cancer(GC)is a considerable global health burden;the median survival of advaneed GC is less than 1 year.^(1)Cancer stem cells(CSCs),a small population of cancer cells with stem cell-like properties,are the major cause of treatme nt failure,in eluding GC,2 however,the mechanisms underlying stemness maintenanee of GC stem cells(GCSCs)are still poorly understood.