GAE-LGA: integration of multi-omics data with graph autoencoders to identify lncRNA-PCG associations.

Long non-coding RNAs (lncRNAs) can disrupt the biological functions of protein-coding genes (PCGs) to cause cancer. However, the relationship between lncRNAs and PCGs remains unclear and difficult to predict. Machine learning has achieved a satisfactory performance in association prediction, but to our knowledge, it is currently less used in lncRNA-PCG association prediction. Therefore, we introduce GAE-LGA, a powerful deep learning model with graph autoencoders as components, to recognize potential lncRNA-PCG associations. GAE-LGA jointly explored lncRNA-PCG learning and cross-omics correlation learning for effective lncRNA-PCG association identification. The functional similarity and multi-omics similarity of lncRNAs and PCGs were accumulated and encoded by graph autoencoders to extract feature representations of lncRNAs and PCGs, which were subsequently used for decoding to obtain candidate lncRNA-PCG pairs. Comprehensive evaluation demonstrated that GAE-LGA can successfully capture lncRNA-PCG associations with strong robustness and outperformed other machine learning-based identification methods. Furthermore, multi-omics features were shown to improve the performance of lncRNA-PCG association identification. In conclusion, GAE-LGA can act as an efficient application for lncRNA-PCG association prediction with the following advantages: It fuses multi-omics information into the similarity network, making the feature representation more accurate; it can predict lncRNA-PCG associations for new lncRNAs and identify potential lncRNA-PCG associations with high accuracy.

Design of Refined Quaternary Electrolyte LiF-LiCl-LiBr-LiI Used for the Liquid Metal Battery.

The advanced electrolyte of liquid metal battery should have low melting point, low ionic solubility, low viscosity, high electric and thermal conductivities, and a suitable density between anode and cathode for declining the operating temperature and realizing the goal of saving-energy. In this study, an excellent quaternary LiF-LiCl-LiBr-LiI (9.1 : 30.0 : 21.7 : 39.2) electrolyte is refined by using thermodynamic models to balance various properties of LiX (X=F, Cl, Br, I) and meet the requirement of advanced electrolyte of liquid metal battery. The refined properties of electrolyte correspond to 2.398 g/cm3 for density, 0.286 mol% for solubility, 4.486 Ohm-1 cm-1 for ionic conductivity, and 0.609 W m-1 for thermal conductivity. The measured melting point is 609.1 K, which is lower than the current operating temperature of 723 K for the lithium-based liquid metal battery. The refined electrolyte consisted by quaternary halide molten-salt provides important references for preparing the advanced liquid metal battery.

Strong Dependence of Thermal and Magnetic Properties on Valence Electron Structures in R2Co17 (R = Rare Earth) Intermetallics.

The valence electron structures (VESs) and thermal and magnetic properties of R2Co17 intermetallics with rhombohedral (R = Ce, Pr, Nd, Sm, Gd, and Tb) and hexagonal (R = Y, Dy, Ho, and Er) structures are studied systematically with the empirical electron theory of solids and molecules (EET). The calculated values, which cover the bond length, cohesive energy, melting point, magnetic moment, and Curie temperature, fit the experimental ones well. The study reveals that the thermal and magnetic properties of R2Co17 are strongly related to their VESs. It shows that the properties of R2Co17 can be modulated by covalence electron number nc/atom for cohesive energy and melting point, the 3d magnetic electrons of various Co sublattices for magnetic moment, the electron transformation from covalence electrons to 3d magnetic electrons for the moments of various Co sublattices, and molecular moment for Curie temperature. The structural stability of R2Co17 depends upon the distribution probability of covalence electrons on various bonds. The pseudobinary La-Co 2:17 phase can be stabilized by doping a transition metal into La2Co17 by modulating the covalence electron number per Co atom to be very close to the stable nc/Co range of rhombohedral LR2Co17 (LR=light rare earth).

Linking genotype to phenotype in multi-omics data of small sample.

BACKGROUND: Genome-wide association studies (GWAS) that link genotype to phenotype represent an effective means to associate an individual genetic background with a disease or trait. However, single-omics data only provide limited information on biological mechanisms, and it is necessary to improve the accuracy for predicting the biological association between genotype and phenotype by integrating multi-omics data. Typically, gene expression data are integrated to analyze the effect of single nucleotide polymorphisms (SNPs) on phenotype. Such multi-omics data integration mainly follows two approaches: multi-staged analysis and meta-dimensional analysis, which respectively ignore intra-omics and inter-omics associations. Moreover, both approaches require omics data from a single sample set, and the large feature set of SNPs necessitates a large sample size for model establishment, but it is difficult to obtain multi-omics data from a single, large sample set. RESULTS: To address this problem, we propose a method of genotype-phenotype association based on multi-omics data from small samples. The workflow of this method includes clustering genes using a protein-protein interaction network and gene expression data, screening gene clusters with group lasso, obtaining SNP clusters corresponding to the selected gene clusters through expression quantitative trait locus data, integrating SNP clusters and corresponding gene clusters and phenotypes into three-layer network blocks, analyzing and predicting based on each block, and obtaining the final prediction by taking the average. CONCLUSIONS: We compare this method to others using two datasets and find that our method shows better results in both cases. Our method can effectively solve the prediction problem in multi-omics data of small sample, and provide valuable resources for further studies on the fusion of more omics data.

Early Death Incidence and Prediction in Stage IV Breast Cancer.

BACKGROUND The early death of patients is a global cancer issue. We aimed to identify the risk factors for early death in stage IV breast cancer. Predictive nomograms for early death evaluation were generated based on the risk factors. MATERIAL AND METHODS Based on the Surveillance, Epidemiology, and End Results (SEER) database, patients diagnosed with IV breast cancer were selected. The risk factors for early death (survival time ≤1 year) were identified using logistic regression model analysis. Predictive nomograms were constructed and internal validation was performed. RESULTS A total of 5998 (32.6%) breast cancer patients were diagnosed as early death in the construction cohort. Age older than 50 years, unmarried status, black race, uninsured status, triple-negative type, grade (II and III), tumor size >5 cm, and metastasis to lung, liver, and brain were risk factors for total early death, while Luminal B subtype, N1 stage, and surgical interventions were associated with lower risk of early death. As for cancer-specific and non-cancer-specific early death, several factors were not consistent between the 2 groups. Nomograms for all-cause, cancer-specific, and non-cancer-specific early death were constructed. The calibration curve showed satisfactory agreement. The areas under the ROC curve (AUC) were 78.3% (95% CI: 77.7-78.9%), 75.8% (75.1-76.4%), and 72.3% (71.6-72.9%), respectively. In the validation cohort, a total of 689 (19.3%) patients were diagnosed as early death and the calibration curve showed satisfactory agreement. The AUCs of the all-cause, cancer-specific, and non-cancer-specific early death prediction were 74.0% (95% CI: 72.5-75.4%), 73.5% (72.0-74.9%), and 68.6% (67.0-70.1%), respectively. CONCLUSIONS Nomograms were generated to predict early death, with good calibration and discrimination. The predictive model can provide a reference for identifying cases with high risk of early death among stage IV breast cancer patients and play an auxiliary role in guiding individual treatment.

Mesoporous sol-gel silica supported vanadium oxide as effective catalysts in oxidative dehydrogenation of propane to propylene.

Vanadium oxide incorporated mesoporous silica (V-m-SiO2) were designed and synthesized using a surfactant-modified sol-gel method. Detailed characterization shows that monomeric [VO4] sites containing one terminal V[double bond, length as m-dash]O bond and three V-O-support bonds are dominated until atomic ratio of vanadium to silicon approaches to 5%. It is also confirmed that such V-m-SiO2 catalyst contains high proportion of vanadium oxide species interacting strongly with silica. Compared to vanadium oxide supported mesoporous silica (V/m-SiO2) prepared using a traditional impregnation method, present V-m-SiO2 catalyst exhibits more superior ability to catalyze oxidative dehydrogenation of propane to propylene. By correlation with structural data, superior catalytic performance of present V-m-SiO2 catalyst can be reasonably attributed, in part, to its favorable geometric and electronic properties rendered by the unique preparation method.

Dietary Restriction and Rapamycin Affect Brain Aging in Mice by Attenuating Age-Related DNA Methylation Changes.

The fact that dietary restriction (DR) and long-term rapamycin treatment (RALL) can ameliorate the aging process has been reported by many researchers. As the interface between external and genetic factors, epigenetic modification such as DNA methylation may have latent effects on the aging rate at the molecular level. To understand the mechanism behind the impacts of dietary restriction and rapamycin on aging, DNA methylation and gene expression changes were measured in the hippocampi of different-aged mice. Examining the single-base resolution of DNA methylation, we discovered that both dietary restriction and rapamycin treatment can maintain DNA methylation in a younger state compared to normal-aged mice. Through functional enrichment analysis of genes in which DNA methylation or gene expression can be affected by DR/RALL, we found that DR/RALL may retard aging through a relationship in which DNA methylation and gene expression work together not only in the same gene but also in the same biological process. This study is instructive for understanding the maintenance of DNA methylation by DR/RALL in the aging process, as well as the role of DR and RALL in the amelioration of aging.

Using expression quantitative trait loci data and graph-embedded neural networks to uncover genotype-phenotype interactions.

Motivation: A central goal of current biology is to establish a complete functional link between the genotype and phenotype, known as the so-called genotype-phenotype map. With the continuous development of high-throughput technology and the decline in sequencing costs, multi-omics analysis has become more widely employed. While this gives us new opportunities to uncover the correlation mechanisms between single-nucleotide polymorphism (SNP), genes, and phenotypes, multi-omics still faces certain challenges, specifically: 1) When the sample size is large enough, the number of omics types is often not large enough to meet the requirements of multi-omics analysis; 2) each omics' internal correlations are often unclear, such as the correlation between genes in genomics; 3) when analyzing a large number of traits (p), the sample size (n) is often smaller than p, n << p, hindering the application of machine learning methods in the classification of disease outcomes. Results: To solve these issues with multi-omics and build a robust classification model, we propose a graph-embedded deep neural network (G-EDNN) based on expression quantitative trait loci (eQTL) data, which achieves sparse connectivity between network layers to prevent overfitting. The correlation within each omics is also considered such that the model more closely resembles biological reality. To verify the capabilities of this method, we conducted experimental analysis using the GSE28127 and GSE95496 data sets from the Gene Expression Omnibus (GEO) database, tested various neural network architectures, and used prior data for feature selection and graph embedding. Results show that the proposed method could achieve a high classification accuracy and easy-to-interpret feature selection. This method represents an extended application of genotype-phenotype association analysis in deep learning networks.

ImReLnc: Identifying Immune-Related LncRNA Characteristics in Human Cancers Based on Heuristic Correlation Optimization.

Long non-coding RNAs (lncRNAs) play critical roles in cancer through gene expression and immune regulation. Identifying immune-related lncRNA (irlncRNA) characteristics would contribute to dissecting the mechanism of cancer pathogenesis. Some computational methods have been proposed to identify irlncRNA characteristics in human cancers, but most of them are aimed at identifying irlncRNA characteristics in specific cancer. Here, we proposed a new method, ImReLnc, to recognize irlncRNA characteristics for 33 human cancers and predict the pathogenicity levels of these irlncRNAs across cancer types. We first calculated the heuristic correlation coefficient between lncRNAs and mRNAs for immune-related enrichment analysis. Especially, we analyzed the relationship between lncRNAs and 17 immune-related pathways in 33 cancers to recognize the irlncRNA characteristics of each cancer. Then, we calculated the Pscore of the irlncRNA characteristics to evaluate their pathogenicity levels. The results showed that highly pathogenic irlncRNAs appeared in a higher proportion of known disease databases and had a significant prognostic effect on cancer. In addition, it was found that the expression of irlncRNAs in immune cells was higher than that of non-irlncRNAs, and the proportion of irlncRNAs related to the levels of immune infiltration was much higher than that of non-irlncRNAs. Overall, ImReLnc accurately identified the irlncRNA characteristics in multiple cancers based on the heuristic correlation coefficient. More importantly, ImReLnc effectively evaluated the pathogenicity levels of irlncRNAs across cancer types. ImReLnc is freely available at https://github.com/meihonggao/ImReLnc.

Magnetic resonance imaging semantic and quantitative features analyses: an additional diagnostic tool for breast phyllodes tumors.

OBJECTIVE: This study aimed to differentiate benign and non-benign (borderline/malignant) phyllodes tumors of the breast by the semantic and quantitative features in magnetic resonance imaging (MRI). METHODS: The female patients, diagnosed with phyllodes tumors by MRI and pathological test, were retrospectively selected from December, 2006 to April, 2019. The MRI of benign, borderline and malignant phyllodes tumors was analyzed using 8 semantic features and 20 computed quantitative features from diffuse contrast-enhanced magnetic resonance imaging (DCE-MRI). The semantic features were analyzed by univariate analysis. The least absolute shrinkage and selection operator (LASSO) method was used to identify the optimal subset of MRI quantitative features. According to the results from multivariate logistic regression for the semantic and quantitative features, the model was constructed to differentiate benign and non-benign (borderline/malignant) phyllodes tumors. RESULTS: Thirty-two benign (58.18%), 13 borderline (23.64%) and 10 malignant (18.18%) phyllodes tumors were identified in 54 patients. Five semantic features were proved to be significantly correlated with pathologic grade, including size, the T1 weighted image signal intensity, fat-saturated T2-weighted image signal intensity, enhanced signal intensity, and kinetic curve pattern. With the analysis of LASSO method, three quantitative texture features with significant predictive ability were selected. The model combining both the semantic and quantitative features was proved to have good performance in differentiation on phyllodes tumors, yielding an area under receiver operating characteristic curve, accuracy, sensitivity and specificity of 0.893, 0.933, 1.000, and 0.818, respectively. CONCLUSION: The constructed model based on the semantic and quantitative features of DCE-MRI can significantly improve the differential diagnosis of phyllodes tumors in breast.

Breast Cancer Molecular Subtype Prediction by Mammographic Radiomic Features.

RATIONALE AND OBJECTIVES: This study aimed to investigate whether quantitative radiomic features extracted from digital mammogram images are associated with molecular subtypes of breast cancer. MATERIALS AND METHODS: In this institutional review board-approved retrospective study, we collected 331 Chinese women who were diagnosed with invasive breast cancer in 2015. This cohort included 29 triple-negative, 45 human epidermal growth factor receptor 2 (HER2)-enriched, 36 luminal A, and 221 luminal B lesions. A set of 39 quantitative radiomic features, including morphologic, grayscale statistic, and texture features, were extracted from the segmented lesion area. Three binary classifications of the subtypes were performed: triple-negative vs non-triple-negative, HER2-enriched vs non-HER2-enriched, and luminal (A + B) vs nonluminal. The Naive Bayes machine learning scheme was employed for the classification, and the least absolute shrink age and selection operator method was used to select the most predictive features for the classifiers. Classification performance was evaluated by the area under receiver operating characteristic curve and accuracy. RESULTS: The model that used the combination of both the craniocaudal and the mediolateral oblique view images achieved the overall best performance than using either of the two views alone, yielding an area under receiver operating characteristic curve (or accuracy) of 0.865 (0.796) for triple-negative vs non-triple-negative, 0.784 (0.748) for HER2-enriched vs non-HER2-enriched, and 0.752 (0.788) for luminal vs nonluminal subtypes. Twelve most predictive features were selected by the least absolute shrink age and selection operator method and four of them (ie, roundness, concavity, gray mean, and correlation) showed a statistical significance (P< .05) in the subtype classification. CONCLUSIONS: Our study showed that quantitative radiomic imaging features of breast tumor extracted from digital mammograms are associated with breast cancer subtypes. Future larger studies are needed to further evaluate the findings.