Biological knowledge graph-guided investigation of immune therapy response in cancer with graph neural network.

The determination of transcriptome profiles that mediate immune therapy in cancer remains a major clinical and biological challenge. Despite responses induced by immune-check points inhibitors (ICIs) in diverse tumor types and all the big breakthroughs in cancer immunotherapy, most patients with solid tumors do not respond to ICI therapies. It still remains a big challenge to predict the ICI treatment response. Here, we propose a framework with multiple prior knowledge networks guided for immune checkpoints inhibitors prediction-DeepOmix-ICI (or ICInet for short). ICInet can predict the immune therapy response by leveraging geometric deep learning and prior biological knowledge graphs of gene-gene interactions. Here, we demonstrate more than 600 ICI-treated patients with ICI response data and gene expression profile to apply on ICInet. ICInet was used for ICI therapy responses prediciton across different cancer types-melanoma, gastric cancer and bladder cancer, which includes 7 cohorts from different data sources. ICInet is able to robustly generalize into multiple cancer types. Moreover, the performance of ICInet in those cancer types can outperform other ICI biomarkers in the clinic. Our model [area under the curve (AUC = 0.85)] generally outperformed other measures, including tumor mutational burden (AUC = 0.62) and programmed cell death ligand-1 score (AUC = 0.74). Therefore, our study presents a prior-knowledge guided deep learning method to effectively select immunotherapy-response-associated biomarkers, thereby improving the prediction of immunotherapy response for precision oncology.

Multi-modality artificial intelligence in digital pathology.

In common medical procedures, the time-consuming and expensive nature of obtaining test results plagues doctors and patients. Digital pathology research allows using computational technologies to manage data, presenting an opportunity to improve the efficiency of diagnosis and treatment. Artificial intelligence (AI) has a great advantage in the data analytics phase. Extensive research has shown that AI algorithms can produce more up-to-date and standardized conclusions for whole slide images. In conjunction with the development of high-throughput sequencing technologies, algorithms can integrate and analyze data from multiple modalities to explore the correspondence between morphological features and gene expression. This review investigates using the most popular image data, hematoxylin-eosin stained tissue slide images, to find a strategic solution for the imbalance of healthcare resources. The article focuses on the role that the development of deep learning technology has in assisting doctors' work and discusses the opportunities and challenges of AI.

ncFANs v2.0: an integrative platform for functional annotation of non-coding RNAs.

Increasing evidence proves the essential regulatory roles of non-coding RNAs (ncRNAs) in biological processes. However, characterizing the specific functions of ncRNAs remains a challenging task, owing to the intensive consumption of the experimental approaches. Here, we present an online platform ncFANs v2.0 that is a significantly enhanced version of our previous ncFANs to provide multiple computational methods for ncRNA functional annotation. Specifically, ncFANs v2.0 was updated to embed three functional modules, including ncFANs-NET, ncFANs-eLnc and ncFANs-CHIP. ncFANs-NET is a new module designed for data-free functional annotation based on four kinds of pre-built networks, including the co-expression network, co-methylation network, long non-coding RNA (lncRNA)-centric regulatory network and random forest-based network. ncFANs-eLnc enables the one-stop identification of enhancer-derived lncRNAs from the de novo assembled transcriptome based on the user-defined or our pre-annotated enhancers. Moreover, ncFANs-CHIP inherits the original functions for microarray data-based functional annotation and supports more chip types. We believe that our ncFANs v2.0 carries sufficient convenience and practicability for biological researchers and facilitates unraveling the regulatory mechanisms of ncRNAs. The ncFANs v2.0 server is freely available at http://bioinfo.org/ncfans or http://ncfans.gene.ac.

NONCODEV6: an updated database dedicated to long non-coding RNA annotation in both animals and plants.

NONCODE (http://www.noncode.org/) is a comprehensive database of collection and annotation of noncoding RNAs, especially long non-coding RNAs (lncRNAs) in animals. NONCODEV6 is dedicated to providing the full scope of lncRNAs across plants and animals. The number of lncRNAs in NONCODEV6 has increased from 548 640 to 644 510 since the last update in 2017. The number of human lncRNAs has increased from 172 216 to 173 112. The number of mouse lncRNAs increased from 131 697 to 131 974. The number of plant lncRNAs is 94 697. The relationship between lncRNAs in human and cancer were updated with transcriptome sequencing profiles. Three important new features were also introduced in NONCODEV6: (i) updated human lncRNA-disease relationships, especially cancer; (ii) lncRNA annotations with tissue expression profiles and predicted function in five common plants; iii) lncRNAs conservation annotation at transcript level for 23 plant species. NONCODEV6 is accessible through http://www.noncode.org/.

KOBAS-i: intelligent prioritization and exploratory visualization of biological functions for gene enrichment analysis.

Gene set enrichment (GSE) analysis plays an essential role in extracting biological insight from genome-scale experiments. ORA (overrepresentation analysis), FCS (functional class scoring), and PT (pathway topology) approaches are three generations of GSE methods along the timeline of development. Previous versions of KOBAS provided services based on just the ORA method. Here we presented version 3.0 of KOBAS, which is named KOBAS-i (short for KOBAS intelligent version). It introduced a novel machine learning-based method we published earlier, CGPS, which incorporates seven FCS tools and two PT tools into a single ensemble score and intelligently prioritizes the relevant biological pathways. In addition, KOBAS has expanded the downstream exploratory visualization for selecting and understanding the enriched results. The tool constructs a novel view of cirFunMap, which presents different enriched terms and their correlations in a landscape. Finally, based on the previous version's framework, KOBAS increased the number of supported species from 1327 to 5944. For an easier local run, it also provides a prebuilt Docker image that requires no installation, as a supplementary to the source code version. KOBAS can be freely accessed at http://kobas.cbi.pku.edu.cn, and a mirror site is available at http://bioinfo.org/kobas.

HERB: a high-throughput experiment- and reference-guided database of traditional Chinese medicine.

Pharmacotranscriptomics has become a powerful approach for evaluating the therapeutic efficacy of drugs and discovering new drug targets. Recently, studies of traditional Chinese medicine (TCM) have increasingly turned to high-throughput transcriptomic screens for molecular effects of herbs/ingredients. And numerous studies have examined gene targets for herbs/ingredients, and link herbs/ingredients to various modern diseases. However, there is currently no systematic database organizing these data for TCM. Therefore, we built HERB, a high-throughput experiment- and reference-guided database of TCM, with its Chinese name as BenCaoZuJian. We re-analyzed 6164 gene expression profiles from 1037 high-throughput experiments evaluating TCM herbs/ingredients, and generated connections between TCM herbs/ingredients and 2837 modern drugs by mapping the comprehensive pharmacotranscriptomics dataset in HERB to CMap, the largest such dataset for modern drugs. Moreover, we manually curated 1241 gene targets and 494 modern diseases for 473 herbs/ingredients from 1966 references published recently, and cross-referenced this novel information to databases containing such data for drugs. Together with database mining and statistical inference, we linked 12 933 targets and 28 212 diseases to 7263 herbs and 49 258 ingredients and provided six pairwise relationships among them in HERB. In summary, HERB will intensively support the modernization of TCM and guide rational modern drug discovery efforts. And it is accessible through http://herb.ac.cn/.

Imbalance of TGF-ß1/BMP-7 pathways induced by M2-polarized macrophages promotes hepatocellular carcinoma aggressiveness.

The transforming growth factor-beta (TGF-ß) signaling pathway is the predominant cytokine signaling pathway in the development and progression of hepatocellular carcinoma (HCC). Bone morphogenetic protein (BMP), another member of the TGF-ß superfamily, has been frequently found to participate in crosstalk with the TGF-ß pathway. However, the complex interaction between the TGF-ß and BMP pathways has not been fully elucidated in HCC. We found that the imbalance of TGF-ß1/BMP-7 pathways was associated with aggressive pathological features and poor clinical outcomes in HCC. The induction of the imbalance of TGF-ß1/BMP-7 pathways in HCC cells could significantly promote HCC cell invasion and stemness by increasing inhibitor of differentiation 1 (ID1) expression. We also found that the microRNA (miR)-17-92 cluster, originating from the extracellular vesicles (EVs) of M2-polarized tumor-associated macrophages (M2-TAMs), stimulated the imbalance of TGF-ß1/BMP-7 pathways in HCC cells by inducing TGF-ß type II receptor (TGFBR2) post-transcriptional silencing and inhibiting activin A receptor type 1 (ACVR1) post-translational ubiquitylation by targeting Smad ubiquitylation regulatory factor 1 (Smurf1). In vivo, short hairpin (sh)-MIR17HG and ACVR1 inhibitors profoundly attenuated HCC cell growth and metastasis by rectifying the imbalance of TGF-ß1/BMP-7 pathways. Therefore, we proposed that the imbalance of TGF-ß1/BMP-7 pathways is a feasible prognostic biomarker and recovering the imbalance of TGF-ß1/BMP-7 pathways might be a potential therapeutic strategy for HCC.

SymMap: an integrative database of traditional Chinese medicine enhanced by symptom mapping.

Recently, the pharmaceutical industry has heavily emphasized phenotypic drug discovery (PDD), which relies primarily on knowledge about phenotype changes associated with diseases. Traditional Chinese medicine (TCM) provides a massive amount of information on natural products and the clinical symptoms they are used to treat, which are the observable disease phenotypes that are crucial for clinical diagnosis and treatment. Curating knowledge of TCM symptoms and their relationships to herbs and diseases will provide both candidate leads and screening directions for evidence-based PDD programs. Therefore, we present SymMap, an integrative database of traditional Chinese medicine enhanced by symptom mapping. We manually curated 1717 TCM symptoms and related them to 499 herbs and 961 symptoms used in modern medicine based on a committee of 17 leading experts practicing TCM. Next, we collected 5235 diseases associated with these symptoms, 19 595 herbal constituents (ingredients) and 4302 target genes, and built a large heterogeneous network containing all of these components. Thus, SymMap integrates TCM with modern medicine in common aspects at both the phenotypic and molecular levels. Furthermore, we inferred all pairwise relationships among SymMap components using statistical tests to give pharmaceutical scientists the ability to rank and filter promising results to guide drug discovery. The SymMap database can be accessed at http://www.symmap.org/ and https://www.bioinfo.org/symmap.

Identification and function annotation of long intervening noncoding RNAs.

RNA-seq technology offers the promise of rapid comprehensive discovery of long intervening noncoding RNAs (lincRNAs). Basic tools such as Tophat and Cufflinks have been widely used for RNA-seq assembly. However, advanced bioinformatics methodologies that allow in-depth analysis of lincRNAs are lacking. Here, we describe a computational protocol that is especially designed for the identification of novel lincRNAs and the prediction of the function. The protocol mainly includes two open-access tools, CNCI and ncFANs. CNCI allows users to distinguish noncoding from protein-coding transcripts and to retrieve novel lincRNAs. ncFANs integrates expression profiles of protein-coding and lincRNA genes to construct coexpression networks. Such networks are subsequently used to perform function predictions of unknown lincRNAs. This protocol will allow users to apply these procedures without the need of additional training. All the tools in current protocol are available http://www.bioinfo.org/np/.

NONCODE 2016: an informative and valuable data source of long non-coding RNAs.

NONCODE (http://www.bioinfo.org/noncode/) is an interactive database that aims to present the most complete collection and annotation of non-coding RNAs, especially long non-coding RNAs (lncRNAs). The recently reduced cost of RNA sequencing has produced an explosion of newly identified data. Revolutionary third-generation sequencing methods have also contributed to more accurate annotations. Accumulative experimental data also provides more comprehensive knowledge of lncRNA functions. In this update, NONCODE has added six new species, bringing the total to 16 species altogether. The lncRNAs in NONCODE have increased from 210 831 to 527,336. For human and mouse, the lncRNA numbers are 167,150 and 130,558, respectively. NONCODE 2016 has also introduced three important new features: (i) conservation annotation; (ii) the relationships between lncRNAs and diseases; and (iii) an interface to choose high-quality datasets through predicted scores, literature support and long-read sequencing method support. NONCODE is also accessible through http://www.noncode.org/.

NONCODEv4: exploring the world of long non-coding RNA genes.

NONCODE (http://www.bioinfo.org/noncode/) is an integrated knowledge database dedicated to non-coding RNAs (excluding tRNAs and rRNAs). Non-coding RNAs (ncRNAs) have been implied in diseases and identified to play important roles in various biological processes. Since NONCODE version 3.0 was released 2 years ago, discovery of novel ncRNAs has been promoted by high-throughput RNA sequencing (RNA-Seq). In this update of NONCODE, we expand the ncRNA data set by collection of newly identified ncRNAs from literature published in the last 2 years and integration of the latest version of RefSeq and Ensembl. Particularly, the number of long non-coding RNA (lncRNA) has increased sharply from 73 327 to 210 831. Owing to similar alternative splicing pattern to mRNAs, the concept of lncRNA genes was put forward to help systematic understanding of lncRNAs. The 56 018 and 46 475 lncRNA genes were generated from 95 135 and 67 628 lncRNAs for human and mouse, respectively. Additionally, we present expression profile of lncRNA genes by graphs based on public RNA-seq data for human and mouse, as well as predict functions of these lncRNA genes. The improvements brought to the database also include an incorporation of an ID conversion tool from RefSeq or Ensembl ID to NONCODE ID and a service of lncRNA identification. NONCODE is also accessible through http://www.noncode.org/.

Phenotype-Genotype Association Analysis of ACTH-Secreting Pituitary Adenoma and Its Molecular Link to Patient Osteoporosis.

Adrenocorticotrophin (ACTH)-secreting pituitary adenoma, also known as Cushing disease (CD), is rare and causes metabolic syndrome, cardiovascular disease and osteoporosis due to hypercortisolism. However, the molecular pathogenesis of CD is still unclear because of a lack of human cell lines and animal models. Here, we study 106 clinical characteristics and gene expression changes from 118 patients, the largest cohort of CD in a single-center. RNA deep sequencing is used to examine genotypic changes in nine paired female ACTH-secreting pituitary adenomas and adjacent nontumorous pituitary tissues (ANPT). We develop a novel analysis linking disease clinical characteristics and whole transcriptomic changes, using Pearson Correlation Coefficient to discover a molecular network mechanism. We report that osteoporosis is distinguished from the phenotype and genotype analysis. A cluster of genes involved in osteoporosis is identified using Pearson correlation coefficient analysis. Most of the genes are reported in the bone related literature, confirming the feasibility of phenotype-genotype association analysis, which could be used in the analysis of almost all diseases. Secreted phosphoprotein 1 (SPP1), collagen type I α 1 chain (COL1A1), 5'-nucleotidase ecto (NT5E), HtrA serine peptidase 1 (HTRA1) and angiopoietin 1 (ANGPT1) and their signalling pathways are shown to be involved in osteoporosis in CD patients. Our discoveries provide a molecular link for osteoporosis in CD patients, and may open new potential avenues for osteoporosis intervention and treatment.

Utilizing sequence intrinsic composition to classify protein-coding and long non-coding transcripts.

It is a challenge to classify protein-coding or non-coding transcripts, especially those re-constructed from high-throughput sequencing data of poorly annotated species. This study developed and evaluated a powerful signature tool, Coding-Non-Coding Index (CNCI), by profiling adjoining nucleotide triplets to effectively distinguish protein-coding and non-coding sequences independent of known annotations. CNCI is effective for classifying incomplete transcripts and sense-antisense pairs. The implementation of CNCI offered highly accurate classification of transcripts assembled from whole-transcriptome sequencing data in a cross-species manner, that demonstrated gene evolutionary divergence between vertebrates, and invertebrates, or between plants, and provided a long non-coding RNA catalog of orangutan. CNCI software is available at http://www.bioinfo.org/software/cnci.

Long non-coding RNAs function annotation: a global prediction method based on bi-colored networks.

More and more evidences demonstrate that the long non-coding RNAs (lncRNAs) play many key roles in diverse biological processes. There is a critical need to annotate the functions of increasing available lncRNAs. In this article, we try to apply a global network-based strategy to tackle this issue for the first time. We develop a bi-colored network based global function predictor, long non-coding RNA global function predictor ('lnc-GFP'), to predict probable functions for lncRNAs at large scale by integrating gene expression data and protein interaction data. The performance of lnc-GFP is evaluated on protein-coding and lncRNA genes. Cross-validation tests on protein-coding genes with known function annotations indicate that our method can achieve a precision up to 95%, with a suitable parameter setting. Among the 1713 lncRNAs in the bi-colored network, the 1625 (94.9%) lncRNAs in the maximum connected component are all functionally characterized. For the lncRNAs expressed in mouse embryo stem cells and neuronal cells, the inferred putative functions by our method highly match those in the known literature.

Identification of prognostic biomarkers in hepatitis B virus-related hepatocellular carcinoma and stratification by integrative multi-omics analysis.

BACKGROUND & AIMS: The differentiation of distinct multifocal hepatocellular carcinoma (HCC): multicentric disease vs. intrahepatic metastases, in which the management and prognosis varies substantively, remains problematic. We aim to stratify multifocal HCC and identify novel diagnostic and prognostic biomarkers by performing whole genome and transcriptome sequencing, as part of a multi-omics strategy. METHODS: A complete collection of tumour and somatic specimens (intrahepatic HCC lesions, matched non-cancerous liver tissue and blood) were obtained from representative patients with multifocal HCC exhibiting two distinct postsurgical courses. Whole-genome and transcriptome sequencing with genotyping were performed for each tissue specimen to contrast genomic alterations, including hepatitis B virus integrations, somatic mutations, copy number variations, and structural variations. We then constructed a phylogenetic tree to visualise individual tumour evolution and performed functional enrichment analyses on select differentially expressed genes to elucidate biological processes involved in multifocal HCC development. Multi-omics data were integrated with detailed clinicopathological information to identify HCC biomarkers, which were further validated using a large cohort of HCC patients (n = 174). RESULTS: The multi-omics profiling and tumour biomarkers could successfully distinguish the two multifocal HCC types, while accurately predicting clonality and aggressiveness. The dual-specificity protein kinase TTK, which is a key mitotic checkpoint regulator with links to p53 signaling, was further shown to be a promising overall prognostic marker for HCC in the large patient cohort. CONCLUSIONS: Comprehensive multi-omics characterisation of multifocal tumour evolution may improve clinical decision-making, facilitate personalised medicine, and expedite identification of novel biomarkers and therapeutic targets in HCC.

Downregulation of Insulin-like growth factor binding protein 6 is associated with ACTH-secreting pituitary adenoma growth.

BACKGROUND: Adrenocorticotrophic hormone (ACTH)-dependent Cushing's syndrome, called Cushing disease, is caused by a corticotroph tumor of the pituitary gland. Insulin-like growth factor binding protein 6 (IGFBP6), which regulates insulin-like growth factor (IGF) activity and inhibits several IGF2-dependent cancer growths, plays a pivotal role in the tumorigenesis of malignancy, but its roles in ACTH-secreting pituitary adenomas remain unclear. OBJECTIVE: To investigate IGFBP6 expression in ACTH-secreting pituitary adenomas, and its involvement in tumor growth. METHODS: Sporadic ACTH-secreting pituitary adenomas specimens (n = 41) and adjacent non-tumorous pituitary tissues (n = 9) were collected by transphenoidal surgery. IGFBP6 expression was assessed by quantitative reverse transcriptase polymerase chain reaction (qRT-PCR) and validated by Western blotting. Associations of IGFBP6 expression with maximum tumor diameter or Ki-67 labeling index were evaluated in ACTH-secreting pituitary adenomas. RESULTS: IGFBP6 mRNA and protein expression were both decreased in ACTH-secreting pituitary adenomas, compared to adjacent non-tumorous pituitary tissues (P < 0.01). IGFBP6 expression was correlated inversely with maximum tumor diameter (Rho = -0.53, P < 0.0001) and Ki-67 levels (Rho = -0.52, P < 0.05). Moreover, IGFBP6 downregulation activated PI3 K-AKT-mTOR pathway in ACTH-secreting pituitary adenomas. CONCLUSIONS: IGFBP6 attenuation in ACTH-secreting pituitary adenomas is associated with tumor growth, through activation of PI3K-AKT-mTOR pathway. The finding underlies IGFBP6 roles in Cushing disease and would potentially provide a novel target of medical therapies.

NONCODE v3.0: integrative annotation of long noncoding RNAs.

Facilitated by the rapid progress of high-throughput sequencing technology, a large number of long noncoding RNAs (lncRNAs) have been identified in mammalian transcriptomes over the past few years. LncRNAs have been shown to play key roles in various biological processes such as imprinting control, circuitry controlling pluripotency and differentiation, immune responses and chromosome dynamics. Notably, a growing number of lncRNAs have been implicated in disease etiology. With the increasing number of published lncRNA studies, the experimental data on lncRNAs (e.g. expression profiles, molecular features and biological functions) have accumulated rapidly. In order to enable a systematic compilation and integration of this information, we have updated the NONCODE database (http://www.noncode.org) to version 3.0 to include the first integrated collection of expression and functional lncRNA data obtained from re-annotated microarray studies in a single database. NONCODE has a user-friendly interface with a variety of search or browse options, a local Genome Browser for visualization and a BLAST server for sequence-alignment search. In addition, NONCODE provides a platform for the ongoing collation of ncRNAs reported in the literature. All data in NONCODE are open to users, and can be downloaded through the website or obtained through the SOAP API and DAS services.

DeepFusion: A deep bimodal information fusion network for unraveling protein-RNA interactions using in vivo RNA structures.

RNA-binding proteins (RBPs) are key post-transcriptional regulators, and the malfunctions of RBP-RNA binding lead to diverse human diseases. However, prediction of RBP binding sites is largely based on RNA sequence features, whereas in vivo RNA structural features based on high-throughput sequencing are rarely incorporated. Here, we designed a deep bimodal information fusion network called DeepFusion for unraveling protein-RNA interactions by incorporating structural features derived from DMS-seq data. DeepFusion integrates two sub-models to extract local motif-like information and long-term context information. We show that DeepFusion performs best compared with other cutting-edge methods with only sequence inputs on two datasets. DeepFusion's performance is further improved with bimodal input after adding in vivo DMS-seq structural features. Furthermore, DeepFusion can be used for analyzing RNA degradation, demonstrating significantly different RBP-binding scores in genes with slow degradation rates versus those with rapid degradation rates. DeepFusion thus provides enhanced abilities for further analysis of functional RNAs. DeepFusion's code and data are available at http://bioinfo.org/deepfusion/.

Optical Genome Mapping Reveals the Landscape of Structural Variations and Their Clinical Significance in HBOC-Related Breast Cancer.

BACKGROUND: Structural variations (SVs) are common genetic alterations in the human genome. However, the profile and clinical relevance of SVs in patients with hereditary breast and ovarian cancer (HBOC) syndrome (germline BRCA1/2 mutations) remains to be fully elucidated. METHODS: Twenty HBOC-related cancer samples (5 breast and 15 ovarian cancers) were studied by optical genome mapping (OGM) and next-generation sequencing (NGS) assays. RESULTS: The SV landscape in the 5 HBOC-related breast cancer samples was comprehensively investigated to determine the impact of intratumor SV heterogeneity on clinicopathological features and on the pattern of genetic alteration. SVs and copy number variations (CNVs) were common genetic events in HBOC-related breast cancer, with a median of 212 SVs and 107 CNVs per sample. The most frequently detected type of SV was insertion, followed by deletion. The 5 HBOC-related breast cancer samples were divided into SVhigh and SVlow groups according to the intratumor heterogeneity of SVs. SVhigh tumors were associated with higher Ki-67 expression, higher homologous recombination deficiency (HRD) scores, more mutated genes, and altered signaling pathways. Moreover, 60% of the HBOC-related breast cancer samples displayed chromothripsis, and 8 novel gene fusion events were identified by OGM and validated by transcriptome data. CONCLUSIONS: These findings suggest that OGM is a promising tool for the detection of SVs and CNVs in HBOC-related breast cancer. Furthermore, OGM can efficiently characterize chromothripsis events and novel gene fusions. SVhigh HBOC-related breast cancers were associated with unfavorable clinicopathological features. SVs may therefore have predictive and therapeutic significance for HBOC-related breast cancers in the clinic.