Meta-analytic principal component analysis in integrative omics application

Batch-effect correction in single-cell RNA sequencing data using JIVE

Motivation

In single-cell RNA sequencing analysis, addressing batch effects-technical artifacts stemming from factors such as varying sequencing technologies, equipment, and capture times-is crucial. These factors can cause unwanted variation and obfuscate the underlying biological signal of interest. The joint and individual variation explained (JIVE) method can be used to extract shared biological patterns from multi-source sequencing data while adjusting for individual non-biological variations (i.e. batch effect). However, its current implementation is originally designed for bulk sequencing data, making it computationally infeasible for large-scale single-cell sequencing datasets.

Results

In this study, we enhance JIVE for large-scale single-cell data by boosting its computational efficiency. Additionally, we introduce a novel application of JIVE for batch-effect correction on multiple single-cell sequencing datasets. Our enhanced method aims to decompose single-cell sequencing datasets into a joint structure capturing the true biological variability and individual structures, which capture technical variability within each batch. This joint structure is then suitable for use in downstream analyses. We benchmarked the results against four popular tools, Seurat v5, Harmony, LIGER, and Combat-seq, which were developed for this purpose. JIVE performed best in terms of preserving cell-type effects and in scenarios in which the batch sizes are balanced.

Availability and implementation

The JIVE implementation used for this analysis can be found at https://github.com/oconnell-statistics-lab/scJIVE.

Expression profiles of miRNAs in the lung tissue of piglets infected with Glaesserella parasuis and the roles of ssc-miR-135 and ssc-miR-155-3p in the regulation of inflammation

MicroRNAs (miRNAs) have been shown to play an important regulatory role in the process of pathogenic infection. However, the miRNAs that regulate the pathogenic process of G. parasuis and their functions are still unknown. Here, high-throughput sequencing was used to quantify the expression of miRNA in piglet lung tissue after G. parasuis XX0306 strain infection. A total of 25 differentially expressed microRNAs (DEmiRNAs) were identified. GO and KEGG pathway enrichment analysis showed that many of the functions of genes that may be regulated by DEmiRNA are related to inflammatory response and immune regulation. Further studies found that ssc-miR-135 may promote the expression of inflammatory factors through NF-κB signaling pathway. Whereas, ssc-miR-155-3p inhibited the inflammatory response induced by G. parasuis, and its regulatory mechanism remains to be further investigated. This study provides a valuable reference for revealing the regulatory effects of miRNAs on the pathogenesis of G. parasuis. DATA AVAILABILITY: The datasets generated during the current study are not publicly available due to this study is currently in the ongoing research stage, and some of the data cannot be made public sooner yet, but are available from the corresponding author on reasonable request.

Review: Cancer and neurodevelopmental disorders: multi-scale reasoning and computational guide

The connection and causality between cancer and neurodevelopmental disorders have been puzzling. How can the same cellular pathways, proteins, and mutations lead to pathologies with vastly different clinical presentations? And why do individuals with neurodevelopmental disorders, such as autism and schizophrenia, face higher chances of cancer emerging throughout their lifetime? Our broad review emphasizes the multi-scale aspect of this type of reasoning. As these examples demonstrate, rather than focusing on a specific organ system or disease, we aim at the new understanding that can be gained. Within this framework, our review calls attention to computational strategies which can be powerful in discovering connections, causalities, predicting clinical outcomes, and are vital for drug discovery. Thus, rather than centering on the clinical features, we draw on the rapidly increasing data on the molecular level, including mutations, isoforms, three-dimensional structures, and expression levels of the respective disease-associated genes. Their integrated analysis, together with chromatin states, can delineate how, despite being connected, neurodevelopmental disorders and cancer differ, and how the same mutations can lead to different clinical symptoms. Here, we seek to uncover the emerging connection between cancer, including pediatric tumors, and neurodevelopmental disorders, and the tantalizing questions that this connection raises.

Principal Component Analysis of a Real-World Cohort of Descemet Stripping Automated Endothelial Keratoplasty and Descemet Membrane Endothelial Keratoplasty Cases: Demonstration of a Powerful Data-Mining Technique for Identifying Areas of Research

PURPOSE

Principal component analysis (PCA) is a descriptive exploratory statistical technique that is widely used in complex fields for data mining. However, it is rarely used in ophthalmology. We explored its research potential with a large series of eyes that underwent 3 keratoplasty techniques: Descemet membrane endothelial keratoplasty (DMEK), conventional Descemet stripping automated endothelial keratoplasty (ConDSAEK), or ultrathin-DSAEK (UT-DSAEK).

METHODS

All consecutive DMEK/DSAEK cases conducted in 2016 to 2022 that had ≥24 months of follow-up were included. ConDSAEK and UT-DSAEK were defined as preoperative central graft thickness ≥130 and <130 μm, respectively. Seventy-six patient, disease, surgical practice, and temporal outcome variables were subjected to PCA, including preoperative anterior keratometry, the use of sulfur hexafluoride gas (SF6) versus air for primary tamponade, and postoperative best corrected visual acuity and endothelial cell density. Associations of interest that were revealed by PCA were assessed with the Welch t test or Pearson test.

RESULTS

A total of 331 eyes were treated with DMEK (n = 165), ConDSAEK (n = 95), or UT-DSAEK (n = 71). PCA showed that ConDSAEK and UT-DSAEK clustered closely, including regarding postoperative best corrected visual acuity, and were clearly distinct from DMEK. PCA and follow-up univariate analyses suggested that in DMEK, 1) flatter preoperative anterior keratometry (average, K1, and K2) associated with more rebubbling (P = 0.004-0.089) and graft detachment (P = 0.007-0.022); 2) graft marking did not affect postoperative ECD; and 3) lower postoperative endothelial cell density associated with SF6 use (all P > 0.001) and longer surgery (P = 0.005-0.091). All associations are currently under additional investigation in our hospital.

CONCLUSIONS

PCA is a powerful technique that can rapidly reveal clinically relevant associations in complex ophthalmological datasets.

Identification and validation of a pyroptosis-related prognostic model for colorectal cancer based on bulk and single-cell RNA sequencing data

BACKGROUND

Pyroptosis impacts the development of malignant tumors, yet its role in colorectal cancer (CRC) prognosis remains uncertain.

AIM

To assess the prognostic significance of pyroptosis-related genes and their association with CRC immune infiltration.

METHODS

Gene expression data were obtained from The Cancer Genome Atlas (TCGA) and single-cell RNA sequencing dataset GSE178341 from the Gene Expression Omnibus (GEO). Pyroptosis-related gene expression in cell clusters was analyzed, and enrichment analysis was conducted. A pyroptosis-related risk model was developed using the LASSO regression algorithm, with prediction accuracy assessed through K-M and receiver operating characteristic analyses. A nomogram predicting survival was created, and the correlation between the risk model and immune infiltration was analyzed using CIBERSORTx calculations. Finally, the differential expression of the 8 prognostic genes between CRC and normal samples was verified by analyzing TCGA-COADREAD data from the UCSC database.

RESULTS

An effective pyroptosis-related risk model was constructed using 8 genes-CHMP2B, SDHB, BST2, UBE2D2, GJA1, AIM2, PDCD6IP, and SEZ6L2 (P < 0.05). Seven of these genes exhibited differential expression between CRC and normal samples based on TCGA database analysis (P < 0.05). Patients with higher risk scores demonstrated increased death risk and reduced overall survival (P < 0.05). Significant differences in immune infiltration were observed between low- and high-risk groups, correlating with pyroptosis-related gene expression.

CONCLUSION

We developed a pyroptosis-related prognostic model for CRC, affirming its correlation with immune infiltration. This model may prove useful for CRC prognostic evaluation.

Prediction of immune infiltration and prognosis for patients with cholangiocarcinoma based on a cuproptosis-related lncRNA signature

Objective

Cholangiocarcinoma (CHOL) is a malignant disease that affects the digestive tract, and it is characterized by a poor prognosis. This research sought to explore the involvement of cuproptosis-related lncRNAs (CRLs) in the prognostic prediction and immune infiltration of cholangiocarcinoma.

Methods

The expression profiles and clinical data of CHOL patients were acquired from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases, and CRLs were defined via co-expression analysis. Two molecular clusters distinguished by cuproptosis-related genes (CRGs) were produced. Then a risk signature consisted by four CRLs was formed, and all samples were separated into low- and high-risk groups using a risk score. Kaplan-Meier survival analysis, principal component analysis, differentially expressed analysis, immune cell infiltration analysis, and sensitivities analysis of chemotherapy drugs were conducted between the two groups. Simultaneously, the expression values of four lncRNAs confirmed by real-time PCR in our own 20 CHOL samples were brought into the risk model.

Results

The CHOL samples could be differentiated into two molecular clusters, which displayed contrasting survival times. Additionally, patients with higher risk scores had significantly worse prognosis compared to those in the low-risk group. Furthermore, both immune infiltration and enrichment analysis revealed significant discrepancies in the tumor immune microenvironment (TIME) between different risk groups. Moreover, the predictive power and the correlation with CA19-9 and CEA of risk signature were validated in our own samples.

Conclusion

We developed a risk signature which could serve as an independent prognostic factor and offer a promising prediction for not only prognosis but also TIME in CHOL patients.

High-throughput sequencing reveals the change of TCR α chain CDR3 with Takayasu arteritis

OBJECTIVE

Takayasu arteritis (TAK) is an inflammatory disease of blood vessels, and its pathogenesis is not clear at present. In this study, we explored the immunological characteristics of T cell receptor (TCR) α-chain complementarity-determining region 3 (CDR3) in patients with TAK.

METHODS

Five untreated patients with TAK were collected from June 2019 to December 2019. Four healthy blood samples were matched as the control group. The blood mononuclear cells were separated, and RNA was extracted for reverse transcription to obtain complementary DNA. Then high-throughput sequencing was performed. The quality of samples was evaluated by principal component analysis. We compared the diversity and expression of TCR α-chain between TAK group and control group. R software was used for statistical analysis and drawing, and Mann-Whitney U test was used to analyze the differences between the two groups.

RESULTS

The results showed that there was a significant difference in the diversity of TCR α-chain CDR3 between the two groups. Three V region genes expression significantly higher in the TAK patients than in the control group. A total of 196 VJ rearrangement genes are significantly different between the two groups, of which 149 rearrangement genes in the TAK group are lower than those in the control group, and 47 rearrangement genes in the TAK group are higher than those in the control group.

CONCLUSION

Patients with TAK have a unique TCR α-chain CDR3 library. These characteristic genes may be a marker for early diagnosis and provide a new theoretical basis for treating TAK.

Integrated single-cell and bulk RNA sequencing analysis identified pyroptosis-related signature for diagnosis and prognosis in osteoarthritis

Osteoarthritis (OA), a degenerative disease of the joints, has one of the highest disability rates worldwide. This study investigates the role of pyroptosis-related genes in osteoarthritis and their expression in different chondrocyte subtypes at the individual cell level. Using OA-related datasets for single-cell RNA sequencing and RNA-seq, the study identified PRDEGs and DEGs and conducted Cox regression analysis to identify independent prognostic factors for OA. CASP6, NOD1, and PYCARD were found to be prognostic factors. Combined Weighted Gene Correlation Network Analysis with PPI network, a total of 15 hub genes related to pyroptosis were involved in the notch and oxidative phosphorylation pathways, which could serve as biomarkers for the diagnosis and prognosis of OA patients. The study also explored the heterogeneity of chondrocytes between OA and normal samples, identifying 19 single-cell subpopulation marker genes that were significantly different among 7 chondrocyte cell clusters. AGT, CTSD, CYBC, and THYS1 were expressed differentially among different cell subpopulations, which were associated with cartilage development and metabolism. These findings provide valuable insights into the molecular mechanisms underlying OA and could facilitate the development of new therapeutic strategies for this debilitating disease.

Single cell RNA sequencing reveals mesenchymal heterogeneity and critical functions of Cd271 in tooth development

BACKGROUND

Accumulating evidence suggests that the maxillary process, to which cranial crest cells migrate, is essential to tooth development. Emerging studies indicate that Cd271 plays an essential role in odontogenesis. However, the underlying mechanisms have yet to be elucidated.

AIM

To establish the functionally heterogeneous population in the maxillary process, elucidate the effects of Cd271 deficiency on gene expression differences.

METHODS

p75NTR knockout (Cd271^-/-) mice (from American Jackson laboratory) were used to collect the maxillofacial process tissue of p75NTR knockout mice, and the wild-type maxillofacial process of the same pregnant mouse wild was used as control. After single cell suspension, the cDNA was prepared by loading the single cell suspension into the 10x Genomics Chromium system to be sequenced by NovaSeq6000 sequencing system. Finally, the sequencing data in Fastq format were obtained. The FastQC software is used to evaluate the quality of data and CellRanger analyzed the data. The gene expression matrix is read by R software, and Seurat is used to control and standardize the data, reduce the dimension and cluster. We search for marker genes for subgroup annotation by consulting literature and database; explore the effect of p75NTR knockout on mesenchymal stem cells (MSCs) gene expression and cell proportion by cell subgrouping, differential gene analysis, enrichment analysis and protein-protein interaction network analysis; understand the interaction between MSCs cells and the differentiation trajectory and gene change characteristics of p75NTR knockout MSCs by cell communication analysis and pseudo-time analysis. Last we verified the findings single cell sequencing in vitro.

RESULTS

We identified 21 cell clusters, and we re-clustered these into three subclusters. Importantly, we revealed the cell–cell communication networks between clusters. We clarified that Cd271 was significantly associated with the regulation of mineralization.

CONCLUSION

This study provides comprehensive mechanistic insights into the maxillary- process-derived MSCs and demonstrates that Cd271 is significantly associated with the odontogenesis in mesenchymal populations.

Identification and validation of fatty acid metabolism-related lncRNA signatures as a novel prognostic model for clear cell renal cell carcinoma

Clear cell renal cell carcinoma (ccRCC) is a main subtype of renal cancer, and advanced ccRCC frequently has poor prognosis. Many studies have found that lipid metabolism influences tumor development and treatment. This study was to examine the prognostic and functional significance of genes associated with lipid metabolism in individuals with ccRCC. Using the database TCGA, differentially expressed genes (DEGs) associated with fatty acid metabolism (FAM) were identified. Prognostic risk score models for genes related to FAM were created using univariate and least absolute shrinkage and selection operator (LASSO) Cox regression analyses. Our findings demonstrate that the prognosis of patients with ccRCC correlate highly with the profiles of FAM-related lncRNAs (AC009166.1, LINC00605, LINC01615, HOXA-AS2, AC103706.1, AC009686.2, AL590094.1, AC093278.2). The prognostic signature can serve as an independent predictive predictor for patients with ccRCC. The predictive signature's diagnostic effectiveness was superior to individual clinicopathological factors. Between the low- and high-risk groups, immunity research revealed a startling difference in terms of cells, function, and checkpoint scores. Chemotherapeutic medications such lapatinib, AZD8055, and WIKI4 had better outcomes for patients in the high-risk group. Overall, the predictive signature can help with clinical selection of immunotherapeutic regimens and chemotherapeutic drugs, improving prognosis prediction for ccRCC patients.

Analysis of cuproptosis-related lncRNA signature for predicting prognosis and tumor immune microenvironment in pancreatic cancer

Pancreatic cancer (PC) is a highly malignant digestive tract tumor, with a dismal 5-year survival rate. Recently, cuproptosis was found to be copper-dependent cell death. This work aims to establish a cuproptosis-related lncRNA signature which could predict the prognosis of PC patients and help clinical decision-making. Firstly, cuproptosis-related lncRNAs were identified in the TCGA-PAAD database. Next, a cuproptosis-related lncRNA signature based on five lncRNAs was established. Besides, the ICGC cohort and our samples from 30 PC patients served as external validation groups to verify the predictive power of the risk signature. Then, the expression of CASC8 was verified in PC samples, scRNA-seq dataset CRA001160, and PC cell lines. The correlation between CASC8 and cuproptosis-related genes was validated by Real-Time PCR. Additionally, the roles of CASC8 in PC progression and immune microenvironment characterization were explored by loss-of-function assay. As showed in the results, the prognosis of patients with higher risk scores was prominently worse than that with lower risk scores. Real-Time PCR and single cell analysis suggested that CASC8 was highly expressed in pancreatic cancer and related to cuproptosis. Additionally, gene inhibition of CASC8 impacted the proliferation, apoptosis and migration of PC cells. Furthermore, CASC8 was demonstrated to impact the expression of CD274 and several chemokines, and serve as a key indicator in tumor immune microenvironment characterization. In conclusion, the cuproptosis-related lncRNA signature could provide valuable indications for the prognosis of PC patients, and CASC8 was a candidate biomarker for not only predicting the progression of PC patients but also their antitumor immune responses.

Microanatomy of the metabolic associated fatty liver disease (MAFLD) by single-cell transcriptomics

BACKGROUND

Metabolic-associated fatty liver disease (MAFLD) is a major cause of liver disease worldwide and comprises non-alcoholic fatty liver (NAFL) and non-alcoholic steatohepatitis (NASH). Due to the high prevalence and poor prognosis of NASH, it is critical to identify and treat patients at risk. However, the aetiology and mechanisms remain largely unknown, warranting further analysis.

METHODS

We first identified differential genes in NASH by single-cell analysis of the GSE129516 dataset and conducted expression profiling data analysis of the GSE184019 dataset from the Gene Expression Omnibus (GEO) database. Then single-cell trajectory reconstruction and analysis, immune gene score, cellular communication, key gene screening, functional enrichment analysis, and immune microenvironment analysis were carried out. Finally, cell experiments were performed to verify the role of key genes in NASH.

RESULTS

We conducted transcriptome profiling of 30,038 single cells, including hepatocytes and non-hepatocytes from normal and steatosis adult mouse livers. Comparative analysis of hepatocytes and non-hepatocytes revealed pronounced heterogeneity as non-hepatocytes acted as major cell-communication hubs. The results showed that Hspa1b, Tfrc, Hmox1 and Map4k4 could effectively distinguish NASH tissues from normal samples. The results of scRNA-seq and qPCR indicated that the expression levels of hub genes in NASH were significantly higher than in normal cells or tissues. Further immune infiltration analysis showed significant differences in M2 macrophage distribution between healthy and metabolic-associated fatty liver samples.

CONCLUSIONS

Our results suggest that Hspa1b, Tfrc, Hmox1 and Map4k4 have huge prospects as diagnostic and prognostic biomarkers for NASH and may be potential therapeutic targets for NASH.

Single-cell RNA sequencing analysis of human chondrocytes reveals cell-cell communication alterations mediated by interactive signaling pathways in osteoarthritis

Objective: Osteoarthritis (OA) is a common joint disorder characterized by degenerative articular cartilage, subchondral bone remodeling, and inflammation. Increasing evidence suggests that the substantial crosstalk between cartilage and synovium is closely related to Osteoarthritis development, but the events that cause this degeneration remain unknown. This study aimed to explore the alterations in intercellular communication involved in the pathogenesis of Osteoarthritis using bioinformatics analysis. Methods: Single-cell transcriptome sequencing (scRNA-seq) profiles derived from articular cartilage tissue of patients with Osteoarthritis were downloaded from a public database. Chondrocyte heterogeneity was assessed using computational analysis, and cell type identification and clustering analysis were performed using the "FindClusters" function in the Seurat package. Intercellular communication networks, including major signaling inputs and outputs for cells, were predicted, and analyzed using CellChat. Results: Seven molecularly defined chondrocytes clusters (homeostatic chondrocytes, hypertrophic chondrocyte (HTC), pre-HTC, regulatory chondrocytes, fibro-chondrocytes (FC), pre-FC, and reparative chondrocyte) with different compositions were identified in the damaged cartilage. Compared to those in the intact cartilage, the overall cell-cell communication frequency and communication strength were remarkably increased in the damaged cartilage. The cellular communication among chondrocyte subtypes mediated by signaling pathways, such as PTN, VISFATIN, SPP1, and TGF-β, was selectively altered in Osteoarthritis. Moreover, we verified that SPP1 pathway enrichment scores increased, but VISFATIN pathway enrichment scores decreased based on the bulk rna-seq datasets in Osteoarthritis. Conclusion: Our results revealed alterations in cell-cell communication among OA-related chondrocyte subtypes that were mediated by specific signaling pathways, which might be a crucial underlying mechanism associated with Osteoarthritis progression.

A multifaceted and feasible prognostic model of amino acid metabolism-related genes in the immune response and tumor microenvironment of head and neck squamous cell carcinomas

We investigated the role of amino acid metabolism (AAM) in head and neck squamous cell carcinoma (HNSCC) tissues to explore its prognostic value and potential therapeutic strategies. A risk score based on four AAM-related genes (AMG) was constructed that could predict the prognosis of HNSCC. These four genes were up-regulated in HNSCC tissues and might act as oncogenes. Internal validation in The Cancer Genome Atlas (TCGA) by bootstrapping showed that patients with high-risk scores had a poorer prognosis than patients with low-risk scores, and this was confirmed in the Gene Expression Omnibus (GEO) cohort. There were also differences between the high-risk and low-risk groups in clinical information and different anatomical sites such as age, sex, TNM stage, grade stage, surgery or no surgery, chemotherapy, radiotherapy, no radiotherapy, neck lymph node dissection or not, and neck lymphovascular invasion, larynx, overlapping lesion of lip, and oral cavity and pharynx tonsil of overall survival (OS). Immune-related characteristics, tumor microenvironment (TME) characteristics, and immunotherapy response were significantly different between high- and low-risk groups. The four AMGs were also found to be associated with the expression of markers of various immune cell subpopulations. Therefore, our comprehensive approach revealed the characterization of AAM in HNSCC to predict prognosis and guide clinical therapy.

A new CCCH-type zinc finger-related lncRNA signature predicts the prognosis of clear cell renal cell carcinoma patients

Background: Clear cell renal cell carcinoma (ccRCC) is the main component of renal cell carcinoma (RCC), and advanced ccRCC frequently indicates a poor prognosis. The significance of the CCCH-type zinc finger (CTZF) gene in cancer has been increasingly demonstrated during the past few years. According to studies, targeted radical therapy for cancer treatment may be a revolutionary therapeutic approach. Both lncRNAs and CCCH-type zinc finger genes are essential in ccRCC. However, the predictive role of long non-coding RNA (lncRNA) associated with the CCCH-type zinc finger gene in ccRCC needs further elucidation. This study aims to predict patient prognosis and investigate the immunological profile of ccRCC patients using CCCH-type zinc finger-associated lncRNAs (CTZFLs).

Methods: From the Cancer Genome Atlas database, RNA-seq and corresponding clinical and prognostic data of ccRCC patients were downloaded. Univariate and multivariate Cox regression analyses were conducted to acquire CTZFLs for constructing prediction models. The risk model was verified using receiver operating characteristic curve analysis. The Kaplan-Meier method was used to analyze the overall survival (OS) of high-risk and low-risk groups. Multivariate Cox and stratified analyses were used to assess the prognostic value of the predictive feature in the entire cohort and different subgroups. In addition, the relationship between risk scores, immunological status, and treatment response was studied.

Results: We constructed a signature consisting of eight CTZFLs (LINC02100, AC002451.1, DBH-AS1, AC105105.3, AL357140.2, LINC00460, DLGAP1-AS2, AL162377.1). The results demonstrated that the prognosis of ccRCC patients was independently predicted by CTZFLs signature and that the prognosis of high-risk groups was poorer than that of the lower group. CTZFLs markers had the highest diagnostic adequacy compared to single clinicopathologic factors, and their AUC (area under the receiver operating characteristic curve) was 0.806. The overall survival of high-risk groups was shorter than that of low-risk groups when patients were divided into groups based on several clinicopathologic factors. There were substantial differences in immunological function, immune cell score, and immune checkpoint expression between high- and low-risk groups. Additionally, Four agents, including ABT737, WIKI4, afuresertib, and GNE 317, were more sensitive in the high-risk group.

Conclusion: The Eight-CTZFLs prognostic signature may be a helpful prognostic indicator and may help with medication selection for clear cell renal cell carcinoma.

Integrative OMICS Data-Driven Procedure Using a Derivatized Meta-Analysis Approach

The wealth of high-throughput data has opened up new opportunities to analyze and describe biological processes at higher resolution, ultimately leading to a significant acceleration of scientific output using high-throughput data from the different omics layers and the generation of databases to store and report raw datasets. The great variability among the techniques and the heterogeneous methodologies used to produce this data have placed meta-analysis methods as one of the approaches of choice to correlate the resultant large-scale datasets from different research groups. Through multi-study meta-analyses, it is possible to generate results with greater statistical power compared to individual analyses. Gene signatures, biomarkers and pathways that provide new insights of a phenotype of interest have been identified by the analysis of large-scale datasets in several fields of science. However, despite all the efforts, a standardized regulation to report large-scale data and to identify the molecular targets and signaling networks is still lacking. Integrative analyses have also been introduced as complementation and augmentation for meta-analysis methodologies to generate novel hypotheses. Currently, there is no universal method established and the different methods available follow different purposes. Herein we describe a new unifying, scalable and straightforward methodology to meta-analyze different omics outputs, but also to integrate the significant outcomes into novel pathways describing biological processes of interest. The significance of using proper molecular identifiers is highlighted as well as the potential to further correlate molecules from different regulatory levels. To show the methodology’s potential, a set of transcriptomic datasets are meta-analyzed as an example.

Identification of a Four Cancer Stem Cell-Related Gene Signature and Establishment of a Prognostic Nomogram Predicting Overall Survival of Pancreatic Adenocarcinoma

BACKGROUND

Cancer stem cells (CSCs) are now being considered as the initial component in the development of pancreatic adenocarcinoma (PAAD). Our aim was to develop a CSCrelated signature to assess the prognosis of PAAD patients for the optimization of treatment.

METHODS

Differentially expressed genes (DEGs) between pancreatic tumor and normal tissue in the Cancer Genome Atlas (TCGA) were screened out, and the weighted gene correlation network analysis (WGCNA) was employed to identify the CSC-related gene sets. Then, univariate, Lasso Cox regression analyses and multivariate Cox regression were applied to construct a prognostic signature using the CSC-related genes. Its prognostic performance was validated in TCGA and ICGC cohorts. Furthermore, Univariate and multivariate Cox regression analyses were used to identify independent prognostic factors in PAAD, and a prognostic nomogram was established.

RESULTS

The Kaplan-Meier analysis, ROC curve and C-index indicated the good performance of the CSC-related signature at predicting overall survival (OS). Univariate Cox regression and multivariate Cox regression revealed that the CSC-related signature was an independent prognostic factor in PAAD. The nomogram was superior to the risk model and AJCC stage in predicting OS. In terms of mutation and tumor immunity, patients in the high-risk group had higher tumor mutation burden (TMB) scores than patients in the low-risk group, and the immune score and the ESTIMATE score were significantly lower in the high-risk group. Moreover, according to the results of principal component analysis (PCA) and Gene Set Enrichment Analysis (GSEA), the low-risk and high-risk groups displayed different stemness statuses based on the risk model.

CONCLUSION

Our study identified four CSC-related gene signatures and established a prognostic nomogram that reliably predicts OS in PAAD. The findings may support new ideas for screening therapeutic targets to inhibit stem characteristics and the development of PAAD.

Identifying Molecular Biomarkers for Diseases With Machine Learning Based on Integrative Omics

Molecular biomarkers are certain molecules or set of molecules that can be of help for diagnosis or prognosis of diseases or disorders. In the past decades, thanks to the advances in high-throughput technologies, a huge amount of molecular 'omics' data, e.g., transcriptomics and proteomics, have been accumulated. The availability of these omics data makes it possible to screen biomarkers for diseases or disorders. Accordingly, a number of computational approaches have been developed to identify biomarkers by exploring the omics data. In this review, we present a comprehensive survey on the recent progress of identification of molecular biomarkers with machine learning approaches. Specifically, we categorize the machine learning approaches into supervised, un-supervised and recommendation approaches, where the biomarkers including single genes, gene sets and small gene networks. In addition, we further discuss potential problems underlying bio-medical data that may pose challenges for machine learning, and provide possible directions for future biomarker identification.

PERTURBED FACTOR ANALYSIS: ACCOUNTING FOR GROUP DIFFERENCES IN EXPOSURE PROFILES

In this article we investigate group differences in phthalate exposure profiles using NHANES data. Phthalates are a family of industrial chemicals used in plastics and as solvents. There is increasing evidence of adverse health effects of exposure to phthalates on reproduction and neurodevelopment and concern about racial disparities in exposure. We would like to identify a single set of low-dimensional factors summarizing exposure to different chemicals, while allowing differences across groups. Improving on current multigroup additive factor models, we propose a class of Perturbed Factor Analysis (PFA) models that assume a common factor structure after perturbing the data via multiplication by a group-specific matrix. Bayesian inference algorithms are defined using a matrix normal hierarchical model for the perturbation matrices. The resulting model is just as flexible as current approaches in allowing arbitrarily large differences across groups but has substantial advantages that we illustrate in simulation studies. Applying PFA to NHANES data, we learn common factors summarizing exposures to phthalates, while showing clear differences across groups.

Identification of a glycolysis-related lncRNA prognostic signature for clear cell renal cell carcinoma

Background: The present study investigated the independent prognostic value of glycolysis-related long noncoding (lnc)RNAs in clear cell renal cell carcinoma (ccRCC).

Methods: A coexpression analysis of glycolysis-related mRNAs–long noncoding RNAs (lncRNAs) in ccRCC from The Cancer Genome Atlas (TCGA) was carried out. Clinical samples were randomly divided into training and validation sets. Univariate Cox regression and least absolute shrinkage and selection operator (LASSO) regression analyses were performed to establish a glycolysis risk model with prognostic value for ccRCC, which was validated in the training and validation sets and in the whole cohort by Kaplan–Meier, univariate and multivariate Cox regression, and receiver operating characteristic (ROC) curve analyses. Principal component analysis (PCA) and functional annotation by gene set enrichment analysis (GSEA) were performed to evaluate the risk model.

Results: We identified 297 glycolysis-associated lncRNAs in ccRCC; of these, 7 were found to have prognostic value in ccRCC patients by Kaplan–Meier, univariate and multivariate Cox regression, and ROC curve analyses. The results of the GSEA suggested a close association between the 7-lncRNA signature and glycolysis-related biological processes and pathways.

Conclusion: The seven identified glycolysis-related lncRNAs constitute an lncRNA signature with prognostic value for ccRCC and provide potential therapeutic targets for the treatment of ccRCC patients.

Identification of pathological transcription in autosomal dominant polycystic kidney disease epithelia

Autosomal dominant polycystic kidney disease (ADPKD) affects more than 12 million people worldwide. Mutations in PKD1 and PKD2 cause cyst formation through unknown mechanisms. To unravel the pathogenic mechanisms in ADPKD, multiple studies have investigated transcriptional mis-regulation in cystic kidneys from patients and mouse models, and numerous dysregulated genes and pathways have been described. Yet, the concordance between studies has been rather limited. Furthermore, the cellular and genetic diversity in cystic kidneys has hampered the identification of mis-expressed genes in kidney epithelial cells with homozygous PKD mutations, which are critical to identify polycystin-dependent pathways. Here we performed transcriptomic analyses of Pkd1- and Pkd2-deficient mIMCD3 kidney epithelial cells followed by a meta-analysis to integrate all published ADPKD transcriptomic data sets. Based on the hypothesis that Pkd1 and Pkd2 operate in a common pathway, we first determined transcripts that are differentially regulated by both genes. RNA sequencing of genome-edited ADPKD kidney epithelial cells identified 178 genes that are concordantly regulated by Pkd1 and Pkd2. Subsequent integration of existing transcriptomic studies confirmed 31 previously described genes and identified 61 novel genes regulated by Pkd1 and Pkd2. Cluster analyses then linked Pkd1 and Pkd2 to mRNA splicing, specific factors of epithelial mesenchymal transition, post-translational protein modification and epithelial cell differentiation, including CD34, CDH2, CSF2RA, DLX5, HOXC9, PIK3R1, PLCB1 and TLR6. Taken together, this model-based integrative analysis of transcriptomic alterations in ADPKD annotated a conserved core transcriptomic profile and identified novel candidate genes for further experimental studies.

Biomarker Categorization in Transcriptomic Meta-Analysis by Concordant Patterns With Application to Pan-Cancer Studies

With the increasing availability and dropping cost of high-throughput technology in recent years, many-omics datasets have accumulated in the public domain. Combining multiple transcriptomic studies on related hypothesis via meta-analysis can improve statistical power and reproducibility over single studies. For differential expression (DE) analysis, biomarker categorization by DE pattern across studies is a natural but critical task following biomarker detection to help explain between study heterogeneity and classify biomarkers into categories with potentially related functionality. In this paper, we propose a novel meta-analysis method to categorize biomarkers by simultaneously considering the concordant pattern and the biological and statistical significance across studies. Biomarkers with the same DE pattern can be analyzed together in downstream pathway enrichment analysis. In the presence of different types of transcripts (e.g., mRNA, miRNA, and lncRNA, etc.), integrative analysis including miRNA/lncRNA target enrichment analysis and miRNA-mRNA and lncRNA-mRNA causal regulatory network analysis can be conducted jointly on all the transcripts of the same category. We applied our method to two Pan-cancer transcriptomic study examples with single or multiple types of transcripts available. Targeted downstream analysis identified categories of biomarkers with unique functionality and regulatory relationships that motivate new hypothesis in Pan-cancer analysis.

The Effect of RBP4 on microRNA Expression Profiles in Porcine Granulosa Cells

Simple Summary

Retinol binding protein 4 (RBP4), mainly secreted by the liver and adipocytes, is a transporter of vitamin A. RBP4 has been shown to be involved in several pathophysiological processes, such as polycystic ovary syndrome (PCOS), obesity, insulin resistance, and cardiovascular risk. However, the role of RBP4 in mammalian follicular granulosa cells (GCs) remains largely unknown. To characterize the molecular pathways associated with the effects of RBP4 on GCs, we used sRNA deep sequencing to detect differential microRNA (miRNA) expression in GCs overexpressing RBP4. A total of 17 miRNAs were significantly different between the experimental and control groups. Our results support the notion that several miRNAs are involved in important biological processes associated with folliculogenesis and pathogenesis. These results will be useful for further studies investigating the role of RBP4 in porcine GCs.

Abstract

Retinol binding protein 4 (RBP4) is a transporter of vitamin A that is secreted mainly by hepatocytes and adipocytes. It affects diverse pathophysiological processes, such as obesity, insulin resistance, and cardiovascular diseases. MicroRNAs (miRNAs) have been reported to play indispensable roles in regulating various developmental processes via the post-transcriptional repression of target genes in mammals. However, the functional link between RBP4 and changes in miRNA expression in porcine granulosa cells (GCs) remains to be investigated. To examine how increased expression of RBP4 affects miRNA expression, porcine GCs were infected with RBP4-targeted lentivirus for 72 h, and whole-genome miRNA profiling (miRNA sequencing) was performed. The sequencing data were validated using real-time quantitative polymerase chain reaction (RT-qPCR) analysis. As a result, we obtained 2783 known and 776 novel miRNAs. In the experimental group, 10 and seven miRNAs were significantly downregulated and upregulated, respectively, compared with the control group. Ontology analysis of the biological processes of these miRNAs indicated their involvement in a variety of biological functions. Gene Ontology and Kyoto Encyclopedia of Genes and Genomes enrichment analyses indicated that these miRNAs were involved mainly in the chemokine signaling pathway, peroxisome proliferators-activated receptors (PPAR) signaling pathway, insulin resistance pathway, nuclear factor-kappa B(NF-kappa B) signaling pathway, and steroid hormone biosynthesis. Our results indicate that RBP4 can regulate the expression of miRNAs in porcine GCs, with consequent physiological effects. In summary, this study profiling miRNA expression in RBP4-overexpressing porcine GCs provides an important reference point for future studies on the regulatory roles of miRNAs in the porcine reproductive system.

Incorporating Machine Learning into Established Bioinformatics Frameworks

The exponential growth of biomedical data in recent years has urged the application of numerous machine learning techniques to address emerging problems in biology and clinical research. By enabling the automatic feature extraction, selection, and generation of predictive models, these methods can be used to efficiently study complex biological systems. Machine learning techniques are frequently integrated with bioinformatic methods, as well as curated databases and biological networks, to enhance training and validation, identify the best interpretable features, and enable feature and model investigation. Here, we review recently developed methods that incorporate machine learning within the same framework with techniques from molecular evolution, protein structure analysis, systems biology, and disease genomics. We outline the challenges posed for machine learning, and, in particular, deep learning in biomedicine, and suggest unique opportunities for machine learning techniques integrated with established bioinformatics approaches to overcome some of these challenges.

A novel autophagy-related lncRNA prognostic risk model for breast cancer

Long non-coding RNAs (lncRNAs) are well known as crucial regulators to breast cancer development and are implicated in controlling autophagy. LncRNAs are also emerging as valuable prognostic factors for breast cancer patients. It is critical to identify autophagy-related lncRNAs with prognostic value in breast cancer. In this study, we identified autophagy-related lncRNAs in breast cancer by constructing a co-expression network of autophagy-related mRNAs-lncRNAs from The Cancer Genome Atlas (TCGA). We evaluated the prognostic value of these autophagy-related lncRNAs by univariate and multivariate Cox proportional hazards analyses and eventually obtained a prognostic risk model consisting of 11 autophagy-related lncRNAs (U62317.4, LINC01016, LINC02166, C6orf99, LINC00992, BAIAP2-DT, AC245297.3, AC090912.1, Z68871.1, LINC00578 and LINC01871). The risk model was further validated as a novel independent prognostic factor for breast cancer patients based on the calculated risk score by Kaplan-Meier analysis, univariate and multivariate Cox regression analyses and time-dependent receiver operating characteristic (ROC) curve analysis. Moreover, based on the risk model, the low-risk and high-risk groups displayed different autophagy and oncogenic statues by principal component analysis (PCA) and Gene Set Enrichment Analysis (GSEA) functional annotation. Taken together, these findings suggested that the risk model of the 11 autophagy-related lncRNAs has significant prognostic value for breast cancer and might be autophagy-related therapeutic targets in clinical practice.

Powerful quantifiers for cancer transcriptomics

Every day, investigators find a new link between a form of cancer and a particular alteration in the sequence or/and expression level of a key gene, awarding this gene the title of “biomarker”. The clinician may choose from numerous available panels to assess the type of cancer based on the mutation or expression regulation (“transcriptomic signature”) of “driver” genes. However, cancer is not a “one-gene show” and, together with the alleged biomarker, hundreds other genes are found as mutated or/and regulated in cancer samples. Regardless of the platform, a well-designed transcriptomic study produces three independent features for each gene: Average expression level, expression variability and coordination with expression of each other gene. While the average expression level is used in all studies to identify what genes were up-/down-regulated or turn on/off, the other two features are unfairly ignored. We use all three features to quantify the transcriptomic change during the progression of the disease and recovery in response to a treatment. Data from our published microarray experiments on cancer nodules and surrounding normal tissue from surgically removed tumors prove that the transcriptomic topologies are not only different in histopathologically distinct regions of a tumor but also dynamic and unique for each human being. We show also that the most influential genes in cancer nodules [the Gene Master Regulators (GMRs)] are significantly less influential in the normal tissue. As such, “smart” manipulation of the cancer GMRs expression may selectively kill cancer cells with little consequences on the normal ones. Therefore, we strongly recommend a really personalized approach of cancer medicine and present the experimental procedure and the mathematical algorithm to identify the most legitimate targets (GMRs) for gene therapy.

The effect of baicalin on microRNA expression profiles in porcine aortic vascular endothelial cells infected by Haemophilus parasuis

Glässer's disease, caused by Haemophilus parasuis (H. parasuis), is associated with vascular damage and vascular inflammation in pigs. Therefore, early assessment and treatment are essential to control the inflammatory disorder. MicroRNAs have been shown to be involved in the vascular pathology. Baicalin has important pharmacological functions, including anti-inflammatory, antimicrobial and antioxidant effects. In this study, we investigated the changes of microRNAs in porcine aortic vascular endothelial cells (PAVECs) induced by H. parasuis and the effect of baicalin in this model by utilizing high-throughput sequencing. The results showed that 155 novel microRNAs and 76 differentially expressed microRNAs were identified in all samples. Subsequently, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of the target genes of the differentially expressed microRNAs demonstrated that regulation of actin cytoskeleton, focal adhesion, ECM-receptor interaction, bacterial invasion of epithelial cells, and adherens junction were the most interesting pathways after PAVECs were infected with H. parasuis. In addition, when the PAVECs were pretreated with baicalin, mismatch repair, peroxisome, oxidative phosphorylation, DNA replication, and ABC transporters were the most predominant signaling pathways. STRING analysis showed that most of the target genes of the differentially expressed microRNAs were associated with each other. The expression levels of the differentially expressed microRNAs were negatively co-regulated with their target genes' mRNA following pretreatment with baicalin in the H. parasuis-induced PAVECs using co-expression networks analysis. This is the first report that microRNAs might have key roles in inflammatory damage of vascular tissue during H. parasuis infection. Baicalin regulated the microRNAs changes in the PAVECs following H. parasuis infection, which may represent useful novel targets to prevent or treat H. parasuis infection.

Identification and validation of stemness-related lncRNA prognostic signature for breast cancer

BACKGROUND

Long noncoding RNAs (lncRNAs) are emerging as crucial contributors to the development of breast cancer and are involved in the stemness regulation of breast cancer stem cells (BCSCs). LncRNAs are closely associated with the prognosis of breast cancer patients. It is critical to identify BCSC-related lncRNAs with prognostic value in breast cancer.

METHODS

A co-expression network of BCSC-related mRNAs-lncRNAs from The Cancer Genome Atlas (TCGA) was constructed. Univariate and multivariate Cox proportional hazards analyses were used to identify a stemness risk model with prognostic value. Kaplan-Meier analysis, univariate and multivariate Cox regression analyses and receiver operating characteristic (ROC) curve analysis were performed to validate the risk model. Principal component analysis (PCA) and Gene Set Enrichment Analysis (GSEA) functional annotation were conducted to analyze the risk model.

RESULTS

In this study, BCSC-related lncRNAs in breast cancer were identified. We evaluated the prognostic value of these BCSC-related lncRNAs and eventually obtained a prognostic risk model consisting of 12 BCSC-related lncRNAs (Z68871.1, LINC00578, AC097639.1, AP003119.3, AP001207.3, LINC00668, AL122010.1, AC245297.3, LINC01871, AP000851.2, AC022509.2 and SEMA3B-AS1). The risk model was further verified as a novel independent prognostic factor for breast cancer patients based on the calculated risk score. Moreover, based on the risk model, the low- risk and high-risk groups displayed different stemness statuses.

CONCLUSIONS

These findings suggested that the 12 BCSC-related lncRNA signature might be a promising prognostic factor for breast cancer and can promote the management of BCSC-related therapy in clinical practice.

Cytokine network analysis of immune responses before and after autologous dendritic cell and tumor cell vaccine immunotherapies in a randomized trial

Background

In a randomized phase II trial conducted in patients with metastatic melanoma, patient-specific autologous dendritic cell vaccines (DCV) were associated with longer survival than autologous tumor cell vaccines (TCV). Both vaccines presented antigens from cell-renewing autologous tumor cells. The current analysis was performed to better understand the immune responses induced by these vaccines, and their association with survival.

Methods

110 proteomic markers were measured at a week-0 baseline, 1 week before the first of 3 weekly vaccine injections, and at week-4, 1 week after the third injection. Data was presented as a deviation from normal controls. A two-component principal component (PC) statistical analysis and discriminant analysis were performed on this data set for all patients and for each treatment cohort.

Results

At baseline PC-1 contained 64.4% of the variance and included the majority of cytokines associated with Th1 and Th2 responses, which positively correlated with beta-2-microglobulin (B2M), programmed death protein-1 (PD-1) and transforming growth factor beta (TGFβ1). Results were similar at baseline for both treatment cohorts. After three injections, DCV-treated patients showed correlative grouping among Th1/Th17 cytokines on PC-1, with an inverse correlation with B2M, FAS, and IL-18, and correlations among immunoglobulins in PC-2. TCV-treated patients showed a positive correlation on PC-1 among most of the cytokines and tumor markers B2M and FAS receptor. There were also correlative changes of IL12p40 with both Th1 and Th2 cytokines and TGFβ1. Discriminant analysis provided additional evidence that DCV was associated with innate, Th1/Th17, and Th2 responses while TCV was only associated with innate and Th2 responses.

Conclusions

These analyses confirm that DCV induced a different immune response than that induced by TCV, and these immune responses were associated with improved survival.

Trial registration Clinical trials.gov NCT004936930 retrospectively registered 28 July 2009

Integrative Deep Learning for Identifying Differentially Expressed (DE) Biomarkers

As a large amount of genetic data are accumulated, an effective analytical method and a significant interpretation are required. Recently, various methods of machine learning have emerged to process genetic data. In addition, machine learning analysis tools using statistical models have been proposed. In this study, we propose adding an integrated layer to the deep learning structure, which would enable the effective analysis of genetic data and the discovery of significant biomarkers of diseases. We conducted a simulation study in order to compare the proposed method with metalogistic regression and meta-SVM methods. The objective function with lasso penalty is used for parameter estimation, and the Youden J index is used for model comparison. The simulation results indicate that the proposed method is more robust for the variance of the data than metalogistic regression and meta-SVM methods. We also conducted real data (breast cancer data (TCGA)) analysis. Based on the results of gene set enrichment analysis, we obtained that TCGA multiple omics data involve significantly enriched pathways which contain information related to breast cancer. Therefore, it is expected that the proposed method will be helpful to discover biomarkers.

Mining cancer biology through bioinformatic analysis of proteomic data

Introduction: Discovery proteomics for cancer research generates complex datasets of diagnostic, prognostic, and therapeutic significance in human cancer. With the advent of high-resolution mass spectrometers, able to identify thousands of proteins in complex biological samples, only the application of bioinformatics can lead to the interpretation of data which can be relevant for cancer research. Areas covered: Here, we give an overview of the current bioinformatic tools used in cancer proteomics. Moreover, we describe their applications in cancer proteomics studies of cell lines, serum, and tissues, highlighting recent results and critically evaluating their outcomes. Expert opinion: The use of bioinformatic tools is a fundamental step in order to manage the large amount of proteins (from hundreds to thousands) that can be identified and quantified in a cancer biological samples by proteomics. To handle this challenge and obtain useful data for translational medicine, it is important the combined use of different bioinformatic tools. Moreover, a particular attention to the global experimental design, and the integration of multidisciplinary skills are essential for best setting of tool parameters and best interpretation of bioinformatics output.

Omics-based analyses revealed metabolic responses of Clostridium acetobutylicum to lignocellulose-derived inhibitors furfural, formic acid and phenol stress for butanol fermentation

BACKGROUND

Clostridium acetobutylicum is a model fermentative anaerobe for consolidated bioprocessing of lignocellulose hydrolysates into acetone-butanol-ethanol (ABE). However, the main inhibitors (acids, furans and phenols) ubiquitous in lignocellulose hydrolysates strictly limit the conversion efficiency. Thus, it is essential to understand the underlying mechanisms of lignocellulose hydrolysate inhibitors to identify key industrial bottlenecks that undermine efficient biofuel production. The recently developed omics strategy for intracellular metabolites and protein quantification now allow for the in-depth mapping of strain metabolism and thereby enable the resolution of the underlying mechanisms.

RESULTS

The toxicity of the main inhibitors in lignocellulose hydrolysates against C. acetobutylicum and ABE production was systematically investigated, and the changes in intracellular metabolism were analyzed by metabolomics and proteomics. The toxicity of the main lignocellulose hydrolysate inhibitors at the same dose was ranked as follows: formic acid > phenol > furfural. Metabolomic analysis based on weighted gene coexpression network analysis (WGCNA) revealed that the three inhibitors triggered the stringent response of C. acetobutylicum. Proteomic analysis based on peptide mass spectrometry (MS) supported the above results and provided more comprehensive conclusions. Under the stress of three inhibitors, the metabolites and key enzymes/proteins involved in glycolysis, reductive tricarboxylic acid (TCA) cycle, acetone-butanol synthesis and redox metabolism were lower than those in the control group. Moreover, proteins involved in gluconeogenesis, the oxidative TCA cycle, thiol peroxidase (TPX) for oxidative stress were significantly upregulated, indicating that inhibitor stress induced the stress response and metabolic regulation. In addition, the three inhibitors also showed stress specificity related to fatty acid synthesis, ATP synthesis, nucleic acid metabolism, nicotinic acid metabolism, cell wall synthesis, spore synthesis and flagellum synthesis and so on.

CONCLUSIONS

Integrated omics platforms provide insight into the cellular responses of C. acetobutylicum to cytotoxic inhibitors released during the deconstruction of lignocellulose. This insight allows us to fully improve the strain to adapt to a challenging culture environment, which will prove critical to the industrial efficacy of C. acetobutylicum.