scGRN: a comprehensive single-cell gene regulatory network platform of human and mouse.

Gene regulatory networks (GRNs) are interpretable graph models encompassing the regulatory interactions between transcription factors (TFs) and their downstream target genes. Making sense of the topology and dynamics of GRNs is fundamental to interpreting the mechanisms of disease etiology and translating corresponding findings into novel therapies. Recent advances in single-cell multi-omics techniques have prompted the computational inference of GRNs from single-cell transcriptomic and epigenomic data at an unprecedented resolution. Here, we present scGRN (https://bio.liclab.net/scGRN/), a comprehensive single-cell multi-omics gene regulatory network platform of human and mouse. The current version of scGRN catalogs 237 051 cell type-specific GRNs (62 999 692 TF-target gene pairs), covering 160 tissues/cell lines and 1324 single-cell samples. scGRN is the first resource documenting large-scale cell type-specific GRN information of diverse human and mouse conditions inferred from single-cell multi-omics data. We have implemented multiple online tools for effective GRN analysis, including differential TF-target network analysis, TF enrichment analysis, and pathway downstream analysis. We also provided details about TF binding to promoters, super-enhancers and typical enhancers of target genes in GRNs. Taken together, scGRN is an integrative and useful platform for searching, browsing, analyzing, visualizing and downloading GRNs of interest, enabling insight into the differences in regulatory mechanisms across diverse conditions.

eRNAbase: a comprehensive database for decoding the regulatory eRNAs in human and mouse.

Enhancer RNAs (eRNAs) transcribed from distal active enhancers serve as key regulators in gene transcriptional regulation. The accumulation of eRNAs from multiple sequencing assays has led to an urgent need to comprehensively collect and process these data to illustrate the regulatory landscape of eRNAs. To address this need, we developed the eRNAbase (http://bio.liclab.net/eRNAbase/index.php) to store the massive available resources of human and mouse eRNAs and provide comprehensive annotation and analyses for eRNAs. The current version of eRNAbase cataloged 10 399 928 eRNAs from 1012 samples, including 858 human samples and 154 mouse samples. These eRNAs were first identified and uniformly processed from 14 eRNA-related experiment types manually collected from GEO/SRA and ENCODE. Importantly, the eRNAbase provides detailed and abundant (epi)genetic annotations in eRNA regions, such as super enhancers, enhancers, common single nucleotide polymorphisms, expression quantitative trait loci, transcription factor binding sites, CRISPR/Cas9 target sites, DNase I hypersensitivity sites, chromatin accessibility regions, methylation sites, chromatin interactions regions, topologically associating domains and RNA spatial interactions. Furthermore, the eRNAbase provides users with three novel analyses including eRNA-mediated pathway regulatory analysis, eRNA-based variation interpretation analysis and eRNA-mediated TF-target gene analysis. Hence, eRNAbase is a powerful platform to query, browse and visualize regulatory cues associated with eRNAs.

KnockTF 2.0: a comprehensive gene expression profile database with knockdown/knockout of transcription (co-)factors in multiple species.

Transcription factors (TFs), transcription co-factors (TcoFs) and their target genes perform essential functions in diseases and biological processes. KnockTF 2.0 (http://www.licpathway.net/KnockTF/index.html) aims to provide comprehensive gene expression profile datasets before/after T(co)F knockdown/knockout across multiple tissue/cell types of different species. Compared with KnockTF 1.0, KnockTF 2.0 has the following improvements: (i) Newly added T(co)F knockdown/knockout datasets in mice, Arabidopsis thaliana and Zea mays and also an expanded scale of datasets in humans. Currently, KnockTF 2.0 stores 1468 manually curated RNA-seq and microarray datasets associated with 612 TFs and 172 TcoFs disrupted by different knockdown/knockout techniques, which are 2.5 times larger than those of KnockTF 1.0. (ii) Newly added (epi)genetic annotations for T(co)F target genes in humans and mice, such as super-enhancers, common SNPs, methylation sites and chromatin interactions. (iii) Newly embedded and updated search and analysis tools, including T(co)F Enrichment (GSEA), Pathway Downstream Analysis and Search by Target Gene (BLAST). KnockTF 2.0 is a comprehensive update of KnockTF 1.0, which provides more T(co)F knockdown/knockout datasets and (epi)genetic annotations across multiple species than KnockTF 1.0. KnockTF 2.0 facilitates not only the identification of functional T(co)Fs and target genes but also the investigation of their roles in the physiological and pathological processes.

LncSEA 2.0: an updated platform for long non-coding RNA related sets and enrichment analysis.

Long non-coding RNAs (lncRNAs) possess a wide range of biological functions, and research has demonstrated their significance in regulating major biological processes such as development, differentiation, and immune response. The accelerating accumulation of lncRNA research has greatly expanded our understanding of lncRNA functions. Here, we introduce LncSEA 2.0 (http://bio.liclab.net/LncSEA/index.php), aiming to provide a more comprehensive set of functional lncRNAs and enhanced enrichment analysis capabilities. Compared with LncSEA 1.0, we have made the following improvements: (i) We updated the lncRNA sets for 11 categories and extremely expanded the lncRNA scopes for each set. (ii) We newly introduced 15 functional lncRNA categories from multiple resources. This update not only included a significant amount of downstream regulatory data for lncRNAs, but also covered numerous epigenetic regulatory data sets, including lncRNA-related transcription co-factor binding, chromatin regulator binding, and chromatin interaction data. (iii) We incorporated two new lncRNA set enrichment analysis functions based on GSEA and GSVA. (iv) We adopted the snakemake analysis pipeline to track data processing and analysis. In summary, LncSEA 2.0 offers a more comprehensive collection of lncRNA sets and a greater variety of enrichment analysis modules, assisting researchers in a more comprehensive study of the functional mechanisms of lncRNAs.

scATAC-Ref: a reference of scATAC-seq with known cell labels in multiple species.

Chromatin accessibility profiles at single cell resolution can reveal cell type-specific regulatory programs, help dissect highly specialized cell functions and trace cell origin and evolution. Accurate cell type assignment is critical for effectively gaining biological and pathological insights, but is difficult in scATAC-seq. Hence, by extensively reviewing the literature, we designed scATAC-Ref (https://bio.liclab.net/scATAC-Ref/), a manually curated scATAC-seq database aimed at providing a comprehensive, high-quality source of chromatin accessibility profiles with known cell labels across broad cell types. Currently, scATAC-Ref comprises 1 694 372 cells with known cell labels, across various biological conditions, >400 cell/tissue types and five species. We used uniform system environment and software parameters to perform comprehensive downstream analysis on these chromatin accessibility profiles with known labels, including gene activity score, TF enrichment score, differential chromatin accessibility regions, pathway/GO term enrichment analysis and co-accessibility interactions. The scATAC-Ref also provided a user-friendly interface to query, browse and visualize cell types of interest, thereby providing a valuable resource for exploring epigenetic regulation in different tissues and cell types.

SpatialRef: a reference of spatial omics with known spot annotation.

Spatial omics technologies have enabled the creation of intricate spatial maps that capture molecular features and tissue morphology, providing valuable insights into the spatial associations and functional organization of tissues. Accurate annotation of spot or domain types is essential for downstream spatial omics analyses, but this remains challenging. Therefore, this study aimed to develop a manually curated spatial omics database (SpatialRef, https://bio.liclab.net/spatialref/), to provide comprehensive and high-quality spatial omics data with known spot labels across multiple species. The current version of SpatialRef aggregates >9 million manually annotated spots across 17 Human, Mouse and Drosophila tissue types through extensive review and strict quality control, covering multiple spatial sequencing technologies and >400 spot/domain types from original studies. Furthermore, SpatialRef supports various spatial omics analyses about known spot types, including differentially expressed genes, spatially variable genes, Gene Ontology (GO)/KEGG annotation, spatial communication and spatial trajectories. With a user-friendly interface, SpatialRef facilitates querying, browsing and visualizing, thereby aiding in elucidating the functional relevance of spatial domains within the tissue and uncovering potential biological effects.

SEanalysis 2.0: a comprehensive super-enhancer regulatory network analysis tool for human and mouse.

Super-enhancers (SEs) play an essential regulatory role in various biological processes and diseases through their specific interaction with transcription factors (TFs). Here, we present the release of SEanalysis 2.0 (http://licpathway.net/SEanalysis), an updated version of the SEanalysis web server for the comprehensive analyses of transcriptional regulatory networks formed by SEs, pathways, TFs, and genes. The current version added mouse SEs and further expanded the scale of human SEs, documenting 1 167 518 human SEs from 1739 samples and 550 226 mouse SEs from 931 samples. The SE-related samples in SEanalysis 2.0 were more than five times that in version 1.0, which significantly improved the ability of original SE-related network analyses ('pathway downstream analysis', 'upstream regulatory analysis' and 'genomic region annotation') for understanding context-specific gene regulation. Furthermore, we designed two novel analysis models, 'TF regulatory analysis' and 'Sample comparative analysis' for supporting more comprehensive analyses of SE regulatory networks driven by TFs. Further, the risk SNPs were annotated to the SE regions to provide potential SE-related disease/trait information. Hence, we believe that SEanalysis 2.0 has significantly expanded the data and analytical capabilities of SEs, which helps researchers in an in-depth understanding of the regulatory mechanisms of SEs.

CRdb: a comprehensive resource for deciphering chromatin regulators in human.

Chromatin regulators (CRs) regulate epigenetic patterns on a partial or global scale, playing a critical role in affecting multi-target gene expression. As chromatin immunoprecipitation sequencing (ChIP-seq) data associated with CRs are rapidly accumulating, a comprehensive resource of CRs needs to be built urgently for collecting, integrating, and processing these data, which can provide abundant annotated information on CR upstream and downstream regulatory analyses as well as CR-related analysis functions. This study established an integrative CR resource, named CRdb (http://cr.liclab.net/crdb/), with the aim of curating a large number of available resources for CRs and providing extensive annotations and analyses of CRs to help biological researchers clarify the regulation mechanism and function of CRs. The CRdb database comprised a total of 647 CRs and 2,591 ChIP-seq samples from more than 300 human tissues and cell types. These samples have been manually curated from NCBI GEO/SRA and ENCODE. Importantly, CRdb provided the abundant and detailed genetic annotations in CR-binding regions based on ChIP-seq. Furthermore, CRdb supported various functional annotations and upstream regulatory information on CRs. In particular, it embedded four types of CR regulatory analyses: CR gene set enrichment, CR-binding genomic region annotation, CR-TF co-occupancy analysis, and CR regulatory axis analysis. CRdb is a useful and powerful resource that can help in exploring the potential functions of CRs and their regulatory mechanism in diseases and biological processes.

SEdb 2.0: a comprehensive super-enhancer database of human and mouse.

Super-enhancers (SEs) are cell-specific DNA cis-regulatory elements that can supervise the transcriptional regulation processes of downstream genes. SEdb 2.0 (http://www.licpathway.net/sedb) aims to provide a comprehensive SE resource and annotate their potential roles in gene transcriptions. Compared with SEdb 1.0, we have made the following improvements: (i) Newly added the mouse SEs and expanded the scale of human SEs. SEdb 2.0 contained 1 167 518 SEs from 1739 human H3K27ac chromatin immunoprecipitation sequencing (ChIP-seq) samples and 550 226 SEs from 931 mouse H3K27ac ChIP-seq samples, which was five times that of SEdb 1.0. (ii) Newly added transcription factor binding sites (TFBSs) in SEs identified by TF motifs and TF ChIP-seq data. (iii) Added comprehensive (epi)genetic annotations of SEs, including chromatin accessibility regions, methylation sites, chromatin interaction regions and topologically associating domains (TADs). (iv) Newly embedded and updated search and analysis tools, including 'Search SE by TF-based', 'Differential-Overlapping-SE analysis' and 'SE-based TF-Gene analysis'. (v) Newly provided quality control (QC) metrics for ChIP-seq processing. In summary, SEdb 2.0 is a comprehensive update of SEdb 1.0, which curates more SEs and annotation information than SEdb 1.0. SEdb 2.0 provides a friendly platform for researchers to more comprehensively clarify the important role of SEs in the biological process.

Cross-modal integration of bulk RNA-seq and single-cell RNA sequencing data to reveal T-cell exhaustion in colorectal cancer.

Colorectal cancer (CRC) is a relatively common malignancy clinically and the second leading cause of cancer-related deaths. Recent studies have identified T-cell exhaustion as playing a crucial role in the pathogenesis of CRC. A long-standing challenge in the clinical management of CRC is to understand how T cells function during its progression and metastasis, and whether potential therapeutic targets for CRC treatment can be predicted through T cells. Here, we propose DeepTEX, a multi-omics deep learning approach that integrates cross-model data to investigate the heterogeneity of T-cell exhaustion in CRC. DeepTEX uses a domain adaptation model to align the data distributions from two different modalities and applies a cross-modal knowledge distillation model to predict the heterogeneity of T-cell exhaustion across diverse patients, identifying key functional pathways and genes. DeepTEX offers valuable insights into the application of deep learning in multi-omics, providing crucial data for exploring the stages of T-cell exhaustion associated with CRC and relevant therapeutic targets.

Programmable, Self-Healable, and Photochromic Liquid Crystal Elastomers Based on Dynamic Hindered Urea Bonds for Biomimetic Flowers.

Recently, researchers have been exploring the use of dynamic covalent bonds (DCBs) in the construction of exchangeable liquid crystal elastomers (LCEs) for biomimetic actuators and devices. However, a significant challenge remains in achieving LCEs with both excellent dynamic properties and superior mechanical strength and stability. In this study, a diacrylate-functionalized monomer containing dynamic hindered urea bonds (DA-HUB) is employed to prepare exchangeable LCEs through a self-catalytic Michael addition reaction. By incorporating DA-HUB, the LCE system benefits from DCBs and hydrogen bonding, leading to materials with high mechanical strength and a range of dynamic properties such as programmability, self-healing, and recyclability. Leveraging these characteristics, bilayer LCE actuators with controlled reversible thermal deformation and outstanding dimensional stability are successfully fabricated using a simple welding method. Moreover, a biomimetic triangular plum, inspired by the blooming of flowers, is created to showcase reversible color and shape changes triggered by light and heat. This innovative approach opens new possibilities for the development of biomimetic and smart actuators and devices with multiple functionalities.

GREAP: a comprehensive enrichment analysis software for human genomic regions.

The rapid development of genomic high-throughput sequencing has identified a large number of DNA regulatory elements with abundant epigenetics markers, which promotes the rapid accumulation of functional genomic region data. The comprehensively understanding and research of human functional genomic regions is still a relatively urgent work at present. However, the existing analysis tools lack extensive annotation and enrichment analytical abilities for these regions. Here, we designed a novel software, Genomic Region sets Enrichment Analysis Platform (GREAP), which provides comprehensive region annotation and enrichment analysis capabilities. Currently, GREAP supports 85 370 genomic region reference sets, which cover 634 681 107 regions across 11 different data types, including super enhancers, transcription factors, accessible chromatins, etc. GREAP provides widespread annotation and enrichment analysis of genomic regions. To reflect the significance of enrichment analysis, we used the hypergeometric test and also provided a Locus Overlap Analysis. In summary, GREAP is a powerful platform that provides many types of genomic region sets for users and supports genomic region annotations and enrichment analyses. In addition, we developed a customizable genome browser containing >400 000 000 customizable tracks for visualization. The platform is freely available at http://www.liclab.net/Greap/view/index.

CanMethdb: a database for genome-wide DNA methylation annotation in cancers.

MOTIVATION: DNA methylation within gene body and promoters in cancer cells is well documented. An increasing number of studies showed that cytosine-phosphate-guanine (CpG) sites falling within other regulatory elements could also regulate target gene activation, mainly by affecting transcription factors (TFs) binding in human cancers. This led to the urgent need for comprehensively and effectively collecting distinct cis-regulatory elements and TF-binding sites (TFBS) to annotate DNA methylation regulation. RESULTS: We developed a database (CanMethdb, http://meth.liclab.net/CanMethdb/) that focused on the upstream and downstream annotations for CpG-genes in cancers. This included upstream cis-regulatory elements, especially those involving distal regions to genes, and TFBS annotations for the CpGs and downstream functional annotations for the target genes, computed through integrating abundant DNA methylation and gene expression profiles in diverse cancers. Users could inquire CpG-target gene pairs for a cancer type through inputting a genomic region, a CpG, a gene name, or select hypo/hypermethylated CpG sets. The current version of CanMethdb documented a total of 38 986 060 CpG-target gene pairs (with 6 769 130 unique pairs), involving 385 217 CpGs and 18 044 target genes, abundant cis-regulatory elements and TFs for 33 TCGA cancer types. CanMethdb might help biologists perform in-depth studies of target gene regulations based on DNA methylations in cancer. AVAILABILITY AND IMPLEMENTATION: The main program is available at https://github.com/chunquanlipathway/CanMethdb. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

TcoFBase: a comprehensive database for decoding the regulatory transcription co-factors in human and mouse.

Transcription co-factors (TcoFs) play crucial roles in gene expression regulation by communicating regulatory cues from enhancers to promoters. With the rapid accumulation of TcoF associated chromatin immunoprecipitation sequencing (ChIP-seq) data, the comprehensive collection and integrative analyses of these data are urgently required. Here, we developed the TcoFBase database (http://tcof.liclab.net/TcoFbase), which aimed to document a large number of available resources for mammalian TcoFs and provided annotations and enrichment analyses of TcoFs. TcoFBase curated 2322 TcoFs and 6759 TcoFs associated ChIP-seq data from over 500 tissues/cell types in human and mouse. Importantly, TcoFBase provided detailed and abundant (epi) genetic annotations of ChIP-seq based TcoF binding regions. Furthermore, TcoFBase supported regulatory annotation information and various functional annotations for TcoFs. Meanwhile, TcoFBase embedded five types of TcoF regulatory analyses for users, including TcoF gene set enrichment, TcoF binding genomic region annotation, TcoF regulatory network analysis, TcoF-TF co-occupancy analysis and TcoF regulatory axis analysis. TcoFBase was designed to be a useful resource that will help reveal the potential biological effects of TcoFs and elucidate TcoF-related regulatory mechanisms.

TF-Marker: a comprehensive manually curated database for transcription factors and related markers in specific cell and tissue types in human.

Transcription factors (TFs) play key roles in biological processes and are usually used as cell markers. The emerging importance of TFs and related markers in identifying specific cell types in human diseases increases the need for a comprehensive collection of human TFs and related markers sets. Here, we developed the TF-Marker database (TF-Marker, http://bio.liclab.net/TF-Marker/), aiming to provide cell/tissue-specific TFs and related markers for human. By manually curating thousands of published literature, 5905 entries including information about TFs and related markers were classified into five types according to their functions: (i) TF: TFs which regulate expression of the markers; (ii) T Marker: markers which are regulated by the TF; (iii) I Marker: markers which influence the activity of TFs; (iv) TFMarker: TFs which play roles as markers and (v) TF Pmarker: TFs which play roles as potential markers. The 5905 entries of TF-Marker include 1316 TFs, 1092 T Markers, 473 I Markers, 1600 TFMarkers and 1424 TF Pmarkers, involving 383 cell types and 95 tissue types in human. TF-Marker further provides a user-friendly interface to browse, query and visualize the detailed information about TFs and related markers. We believe TF-Marker will become a valuable resource to understand the regulation patterns of different tissues and cells.

Efficient benzo(a)pyrene degradation by coal gangue-based catalytic material for peroxymonosulfate activation.

A low cost and green peroxymonosulfate (PMS) activation catalyst (CG-Ca-N) was successfully prepared with coal gangue (CG), calcium chloride, and melamine as activator. Under the optimal conditions, the CG-Ca-N can remove 100 % for benzo(a)pyrene (Bap) in an aqueous solution after 20 min and 72.06 % in soil slurry medium within 60 min, which also display excellent reuse ability toward Bap after three times. The removal of Bap is significantly decreased when the initial pH value was greater than 9 and obviously inhibited in the presence of HCO3- or SO42-. The characterization results indicated that the addition of calcium chloride could stabilize and increase the content of pyridinic N during thermal annealing, resulting in the production of •OH, SO4•- and 1O2. Based on electron paramagnetic resonance (EPR) and active radical scavenging experiments, 1O2 could be identified to be the dominant role in Bap degradation. Overall, this work opened a new perspective for the low cost and green PMS catalysts and offered great promise in the practical remediation of organic pollution of groundwater and soil.

Generation of oxygen vacancies in CuO/TiO2 heterojunction induced by diatomite for highly efficient UV-Vis-IR driven photothermal catalytic oxidation of HCHO.

Photothermal catalytic oxidation is a promising and sustainable method for the degradation of indoor formaldehyde (HCHO). However, the excessively high surface temperature of existing photothermal catalysts during catalysis hinders the effective adsorption and degradation of formaldehyde under static conditions. Catalyst loading and oxygen vacancies (OVs) modulation are commonly employed strategies to reduce the photothermal catalytic temperature and enhance the efficiency of photothermal catalytic oxidation. In this work, a p-n type CuO/TiO2 heterojunction is successfully loaded onto diatomite using a wet precipitation method. Under the irradiation of a 300W xenon lamp, the prepared composite material achieved a 100% removal rate of HCHO within 2 h, with a 98% conversion rate to CO2, surpassing the performance of both individual photocatalysts and thermocatalysts. Additionally, by adjusting conditions such as light irradiation and temperature, we have demonstrated that this material exhibits synergistic photothermal catalytic properties. Based on HRTEM, XPS, Raman, and EPR analyses, the introduction of diatomite as a catalyst support was shown to effectively increase the number of OVs. Experimental results, along with O2-TPD, photoelectrochemical characterization, and radical detection, demonstrate that the presence of OVs enhances the oxidative efficiency of both photocatalysis and thermocatalysis, as well as the UV-Vis-IR photothermal catalytic performance. The ternary composite material generates weak hydroxyl (•OH) and superoxide (•O2-) radical under high-temperature with dark conditions, indicating its catalytic oxidation activity under this condition. The increase in temperature and the expansion of the spectral range both enhance the generation of these radicals. In summary, this work demonstrates that the use of diatomite as a support increases the material's specific surface area and OVs content, thereby enhancing adsorption and photothermal catalysis. It elucidates the enhanced catalytic degradation mechanism of this mineral-based photothermal catalyst.

Development and validation of a multi-modality fusion deep learning model for differentiating glioblastoma from solitary brain metastases. / åŒºåˆ†èƒ¶è´¨æ¯ç»†èƒžç˜¤å’Œå•å‘æ€§è„‘è½¬ç§»ç˜¤çš„å¤šæ¨¡æ€èžåˆæ·±åº¦å­¦ä¹ æ¨¡åž‹çš„å¼€å‘å’ŒéªŒè¯.

OBJECTIVES: Glioblastoma (GBM) and brain metastases (BMs) are the two most common malignant brain tumors in adults. Magnetic resonance imaging (MRI) is a commonly used method for screening and evaluating the prognosis of brain tumors, but the specificity and sensitivity of conventional MRI sequences in differential diagnosis of GBM and BMs are limited. In recent years, deep neural network has shown great potential in the realization of diagnostic classification and the establishment of clinical decision support system. This study aims to apply the radiomics features extracted by deep learning techniques to explore the feasibility of accurate preoperative classification for newly diagnosed GBM and solitary brain metastases (SBMs), and to further explore the impact of multimodality data fusion on classification tasks. METHODS: Standard protocol cranial MRI sequence data from 135 newly diagnosed GBM patients and 73 patients with SBMs confirmed by histopathologic or clinical diagnosis were retrospectively analyzed. First, structural T1-weight, T1C-weight, and T2-weight were selected as 3 inputs to the entire model, regions of interest (ROIs) were manually delineated on the registered three modal MR images, and multimodality radiomics features were obtained, dimensions were reduced using a random forest (RF)-based feature selection method, and the importance of each feature was further analyzed. Secondly, we used the method of contrast disentangled to find the shared features and complementary features between different modal features. Finally, the response of each sample to GBM and SBMs was predicted by fusing 2 features from different modalities. RESULTS: The radiomics features using machine learning and the multi-modal fusion method had a good discriminatory ability for GBM and SBMs. Furthermore, compared with single-modal data, the multimodal fusion models using machine learning algorithms such as support vector machine (SVM), Logistic regression, RF, adaptive boosting (AdaBoost), and gradient boosting decision tree (GBDT) achieved significant improvements, with area under the curve (AUC) values of 0.974, 0.978, 0.943, 0.938, and 0.947, respectively; our comparative disentangled multi-modal MR fusion method performs well, and the results of AUC, accuracy (ACC), sensitivity (SEN) and specificity(SPE) in the test set were 0.985, 0.984, 0.900, and 0.990, respectively. Compared with other multi-modal fusion methods, AUC, ACC, and SEN in this study all achieved the best performance. In the ablation experiment to verify the effects of each module component in this study, AUC, ACC, and SEN increased by 1.6%, 10.9% and 15.0%, respectively after 3 loss functions were used simultaneously. CONCLUSIONS: A deep learning-based contrast disentangled multi-modal MR radiomics feature fusion technique helps to improve GBM and SBMs classification accuracy.

TRlnc: a comprehensive database for human transcriptional regulatory information of lncRNAs.

Long noncoding RNAs (lncRNAs) have been proven to play important roles in transcriptional processes and biological functions. With the increasing study of human diseases and biological processes, information in human H3K27ac ChIP-seq, ATAC-seq and DNase-seq datasets is accumulating rapidly, resulting in an urgent need to collect and process data to identify transcriptional regulatory regions of lncRNAs. We therefore developed a comprehensive database for human regulatory information of lncRNAs (TRlnc, http://bio.licpathway.net/TRlnc), which aimed to collect available resources of transcriptional regulatory regions of lncRNAs and to annotate and illustrate their potential roles in the regulation of lncRNAs in a cell type-specific manner. The current version of TRlnc contains 8 683 028 typical enhancers/super-enhancers and 32 348 244 chromatin accessibility regions associated with 91 906 human lncRNAs. These regions are identified from over 900 human H3K27ac ChIP-seq, ATAC-seq and DNase-seq samples. Furthermore, TRlnc provides the detailed genetic and epigenetic annotation information within transcriptional regulatory regions (promoter, enhancer/super-enhancer and chromatin accessibility regions) of lncRNAs, including common SNPs, risk SNPs, eQTLs, linkage disequilibrium SNPs, transcription factors, methylation sites, histone modifications and 3D chromatin interactions. It is anticipated that the use of TRlnc will help users to gain in-depth and useful insights into the transcriptional regulatory mechanisms of lncRNAs.

Optimal design of a 4 × 4 MMI thermal optical switch with trapezoidal air trenches.

Optical switches are important in signal routing and switching. In this paper, a thermal optical switch with trapezoidal air trenches is proposed. The proposed structure consists of two cascaded 4×4 multimode interference (MMI) couplers. The beam propagation method is used to optimize the dimension and analyze the characteristics. Simulation results show excess loss (EL) and insertion loss (IL) are less than 0.14 dB and 0.22 dB at the wavelength of 1550 nm, respectively. Besides, the extinction ratio (ER) is higher than 21 dB. This design has the advantages of small size, low loss, and high flexibility, which is promising for application to all-optical network routing.