Are There Limits in Explainability of Prognostic Biomarkers? Scrutinizing Biological Utility of Established Signatures

Prognostic biomarkers can have an important role in the clinical practice because they allow stratification of patients in terms of predicting the outcome of a disorder. Obstacles for developing such markers include lack of robustness when using different data sets and limited concordance among similar signatures. In this paper, we highlight a new problem that relates to the biological meaning of already established prognostic gene expression signatures. Specifically, it is commonly assumed that prognostic markers provide sensible biological information and molecular explanations about the underlying disorder. However, recent studies on prognostic biomarkers investigating 80 established signatures of breast and prostate cancer demonstrated that this is not the case. We will show that this surprising result is related to the distinction between causal models and predictive models and the obfuscating usage of these models in the biomedical literature. Furthermore, we suggest a falsification procedure for studies aiming to establish a prognostic signature to safeguard against false expectations with respect to biological utility.

Download Full-text

Identification of a lncRNA prognostic signature-related to stem cell index and its significance in colorectal cancer

Future Oncology ◽

10.2217/fon-2020-1163 ◽

2021 ◽

Author(s):

Xiao-Cheng Wang ◽

Ya Liu ◽

Fei-Wu Long ◽

Liang-Ren Liu ◽

Chuan-Wen Fan

Keyword(s):

Colorectal Cancer ◽

Overall Survival ◽

Prognostic Markers ◽

The Cancer Genome Atlas ◽

Prognostic Signature ◽

Free Survival ◽

Cancer Genome Atlas ◽

Bioinformatic Approaches ◽

Cell Phenotypes ◽

The Relationship

Background: The relationship between long noncoding RNAs (lncRNAs) and the mRNA stemness index (mRNAsi) in colorectal cancer (CRC) is still unclear. Materials & methods: The mRNAsi, mRNAsi-related lncRNAs and their clinical significance were analyzed by bioinformatic approaches in The Cancer Genome Atlas (TCGA)-COREAD dataset. Results: mRNAsi was negatively related to pathological features but positively related to overall survival and recurrence-free survival in CRC. A five mRNAsi-related lncRNAs prognostic signature was further developed and showed independent prognostic factors related to overall survival in CRC patients, due to the five mRNAsi-related lncRNAs involved in several pathways of the cancer stem cells and malignant cancer cell phenotypes. Conclusion: The present study highlights the potential roles of mRNAsi-related lncRNAs as alternative prognostic markers.

Download Full-text

Cancer-Specific Immune Prognostic Signature in Solid Tumors and Its Relation to Immune Checkpoint Therapies

Cancers ◽

10.3390/cancers12092476 ◽

2020 ◽

Vol 12 (9) ◽

pp. 2476 ◽

Cited By ~ 1

Author(s):

Shaoli Das ◽

Kevin Camphausen ◽

Uma Shankavaram

Keyword(s):

Immune Function ◽

Solid Tumors ◽

Immune Checkpoint ◽

Data Sets ◽

Cancer Type ◽

Rna Seq ◽

Prognostic Signature ◽

Cell Functions ◽

Cancer Types ◽

Disease Free

To elucidate the role of immune cell infiltration as a prognostic signature in solid tumors, we analyzed immune-function-related genes from four publicly available single-cell RNA-Seq data sets and twenty bulk tumor RNA-Seq data sets from The Cancer Genome Atlas (TCGA). Unsupervised clustering of pan-cancer transcriptomic signature showed two major immune function types: one related to NK-, T-, and B-cell functions and the other related to monocyte, macrophage, dendritic cell, and Toll-like receptor functions. Kaplan–Meier analysis showed differential prognosis of these two groups, dependent on the cancer type. Our analysis of TCGA solid tumors with an elastic net model identified 155 genes associated with disease-free survival in different tumor types with varied influence across different cancer types. With this gene set, we computed cancer-specific prognostic immune score models for individual cancer types that predicted disease-free and overall survival. Validation of our model on available published data of immune checkpoint blockade therapies on melanoma, kidney renal cell carcinoma, non-small cell lung cancer, gastric cancer and bladder cancer confirmed that cancer-specific higher immune scores are associated with response to immunotherapy. Our analysis provides a comprehensive map of cancer-specific immune-related prognostic gene sets that are associated with immunotherapy response.

Download Full-text

Identification and validation of tumor microenvironment-related lncRNA prognostic signature for uveal melanoma

International Journal of Ophthalmology ◽

10.18240/ijo.2021.08.03 ◽

2021 ◽

Vol 14 (8) ◽

pp. 1151-1159

Author(s):

Chen-Lu Liao ◽

◽

Xing-Yu Sun ◽

Qi Zhou ◽

Min Tian ◽

...

Keyword(s):

Prognostic Value ◽

Prognostic Markers ◽

Immune Cell ◽

Cox Regression ◽

Cell Infiltration ◽

Immune Cell Infiltration ◽

Prognostic Signature ◽

Profile Data ◽

Small Molecule Drugs

AIM: To investigate the role of tumor microenvironment (TME)-related long non-coding RNA (lncRNA) in uveal melanoma (UM), probable prognostic signature and potential small molecule drugs using bioinformatics analysis. METHODS: UM expression profile data were downloaded from the Cancer Genome Atlas (TCGA) and bioinformatics methods were used to find prognostic lncRNAs related to UM immune cell infiltration. The gene expression profile data of 80 TCGA specimens were analyzed using the single sample Gene Set Enrichment Analysis (ssGSEA) method, and the immune cell infiltration of a single specimen was evaluated. Finally, the specimens were divided into high and low infiltration groups. The differential expression between the two groups was analyzed using the R package ‘edgeR’. Univariate, multivariate and Least Absolute Shrinkage and Selection Operator (LASSO) Cox regression analyses were performed to explore the prognostic value of TME-related lncRNAs. Gene Ontology and Kyoto Encyclopedia of Genes and Genomes (KEGG) functional analyses were also performed. The Connectivity Map (CMap) data set was used to screen molecular drugs that may treat UM. RESULTS: A total of 2393 differentially expressed genes were identified and met the criteria for the low and high immune cell infiltration groups. Univariate Cox analysis of lncRNA genes with differential expression identified 186 genes associated with prognosis. Eight prognostic markers of TME-included lncRNA genes were established as potentially independent prognostic elements. Among 269 differentially expressed lncRNAs, 69 were up-regulated and 200 were down-regulated. Univariate Cox regression analysis of the risk indicators and clinical characteristics of the 8 lncRNA gene constructs showed that age, TNM stage, tumor base diameter, and low and high risk indices had significant prognostic value. We screened the potential small-molecule drugs for UM, including W-13, AH-6809 and Imatinib. CONCLUSION: The prognostic markers identified in this study are reliable biomarkers of UM. This study expands our current understanding of the role of TME-related lncRNAs in UM genesis, which may lay the foundations for future treatment of this disease.

Download Full-text

Development of a Multi-gene-based Immune Prognostic Signature in Ovarian Cancer

10.21203/rs.3.rs-52624/v1 ◽

2020 ◽

Author(s):

tiefeng cao ◽

huimin shen

Keyword(s):

Ovarian Cancer ◽

Network Analysis ◽

Prognostic Biomarkers ◽

Survival Prediction ◽

Immune Gene ◽

Prognostic Signature ◽

Data Set ◽

Cox Regression Analysis ◽

Immune Genes ◽

Immune Related Genes

Abstract Background：Chemotherapeutic resistance is responsible for treatment failure. Immunotherapy is important in ovarian cancer (OC). Systematic exploration of immunogenic landscape and reliable immune gene-based prognostic biomarkers or signature is necessary to be identified. This study aims to identify the immune gene-based prognostic biomarkers and regulatory factors, further to develop an individualized prediction signature.Methods: This study systematically explored the gene expression profiles from RNA-seq data set for The Cancer Genome Atlas (TCGA) ovarian cancer. Differentially expressed and survival-associated immune genes and transcription factors (TFs) were identified using immune genes from ImmPort dataset and TFs from Cistoma database. We developed the prognostic signature based on survival associated immune genes with LASSO (Least absolute shrinkage and selection operator) Cox regression analysis. Further, Network analysis was performed to uncover the potential molecular mechanisms of immune-related genes with the help of computational biology. Results: The prognostic signature, a weighted combination of the 21 immune-related genes, performed moderately in survival prediction with AUC was 0.746, 0.735, and 0.749 for 1, 3, and 5 year overall survival, respectively. Network analysis uncovered the regulatory role of TFs in immune genes. Intriguingly, the prognostic signature reflected infiltration of some immune cell subtypes.Conclusions: We first constructed a signature with 21 immune genes of clinical significance, which showed promising predictive value in the surveillance, prognosis, even immunotherapy response of OC patients.

Download Full-text

Semantic-Based Indexing Approaches for Medical Document Clustering Using Cognitive Search

Advances in Social Networking and Online Communities - Cognitive Social Mining Applications in Data Analytics and Forensics ◽

10.4018/978-1-5225-7522-1.ch003 ◽

2019 ◽

pp. 41-64

Author(s):

Logeswari Shanmugam ◽

Premalatha K.

Keyword(s):

Text Processing ◽

Relevant Information ◽

Biomedical Literature ◽

Data Sets ◽

Original Text ◽

Biomedical Knowledge ◽

Text Data ◽

Natural Language Text ◽

Diverse Data ◽

Medical Document

Biomedical literature is the primary repository of biomedical knowledge in which PubMed is the most absolute database for collecting, organizing and analyzing textual knowledge. The high dimensionality of the natural language text makes the text data quite noisy and sparse in the vector space. Hence, the data preprocessing and feature selection are important processes for the text processing issues. Ontologies select the meaningful terms semantically associated with the concepts from a document to reduce the dimensionality of the original text. In this chapter, semantic-based indexing approaches are proposed with cognitive search which makes use of domain ontology to extract relevant information from big and diverse data sets for users.

Download Full-text

Weighted Gene Coexpression Network Analysis Reveals the Dynamic Transcriptome Regulation and Prognostic Biomarkers of Hepatocellular Carcinoma

Evolutionary Bioinformatics ◽

10.1177/1176934320920562 ◽

2020 ◽

Vol 16 ◽

pp. 117693432092056

Author(s):

Shuping Qu ◽

Qiuyuan Shi ◽

Jing Xu ◽

Wanwan Yi ◽

Hengwei Fan

Keyword(s):

Hepatocellular Carcinoma ◽

Differentially Expressed ◽

Prognostic Biomarkers ◽

Coexpression Network ◽

Data Sets ◽

Independent Data ◽

Gene Coexpression Network ◽

Gene Coexpression ◽

Gene Modules ◽

Coexpression Network Analysis

This study was aimed at revealing the dynamic regulation of mRNAs, long noncoding RNAs (lncRNAs), and microRNAs (miRNAs) in hepatocellular carcinoma (HCC) and to identify HCC biomarkers capable of predicting prognosis. Differentially expressed mRNAs (DEmRNAs), lncRNAs, and miRNAs were acquired by comparing expression profiles of HCC with normal samples, using an expression data set from The Cancer Genome Atlas. Altered biological functions and pathways in HCC were analyzed by subjecting DEmRNAs to Gene Ontology and Kyoto Encyclopedia of Genes and Genomes analysis. Gene modules significantly associated with disease status were identified by weighted gene coexpression network analysis. An lncRNA-mRNA and an miRNA-mRNA coexpression network were constructed for genes in disease-related modules, followed by the identification of prognostic biomarkers using Kaplan-Meier survival analysis. Differential expression and association with the prognosis of 4 miRNAs were verified in independent data sets. A total of 1220 differentially expressed genes were identified between HCC and normal samples. Differentially expressed mRNAs were significantly enriched in functions and pathways related to “plasma membrane structure,” “sensory perception,” “metabolism,” and “cell proliferation.” Two disease-associated gene modules were identified. Among genes in lncRNA-mRNA and miRNA-mRNA coexpression networks, 9 DEmRNAs and 7 DEmiRNAs were identified to be potential prognostic biomarkers. MIMAT0000102, MIMAT0003882, and MIMAT0004677 were successfully validated in independent data sets. Our results may advance our understanding of molecular mechanisms underlying HCC. The biomarkers may contribute to diagnosis in future clinical practice.

Download Full-text

Relevance of MicroRNAs as Potential Diagnostic and Prognostic Markers in Colorectal Cancer

International Journal of Molecular Sciences ◽

10.3390/ijms19102944 ◽

2018 ◽

Vol 19 (10) ◽

pp. 2944 ◽

Cited By ~ 23

Author(s):

Grzegorz Hibner ◽

Małgorzata Kimsa-Furdzik ◽

Tomasz Francuz

Keyword(s):

Colorectal Cancer ◽

Prognostic Markers ◽

Expression Profiles ◽

Developed Countries ◽

Prognostic Biomarkers ◽

Screening Methods ◽

Diagnostic Biomarkers ◽

Cellular Processes ◽

Mirna Expression Profiles ◽

And Migration

Colorectal cancer (CRC) is currently the third and the second most common cancer in men and in women, respectively. Every year, more than one million new CRC cases and more than half a million deaths are reported worldwide. The majority of new cases occur in developed countries. Current screening methods have significant limitations. Therefore, a lot of scientific effort is put into the development of new diagnostic biomarkers of CRC. Currently used prognostic markers are also limited in assessing the effectiveness of CRC therapy. MicroRNAs (miRNAs) are a promising subject of research especially since single miRNA can recognize a variety of different mRNA transcripts. MiRNAs have important roles in epigenetic regulation of basic cellular processes, such as proliferation, apoptosis, differentiation, and migration, and may serve as potential oncogenes or tumor suppressors during cancer development. Indeed, in a large variety of human tumors, including CRC, significant distortions in miRNA expression profiles have been observed. Thus, the use of miRNAs as diagnostic and prognostic biomarkers in cancer, particularly in CRC, appears to be an inevitable consequence of the advancement in oncology and gastroenterology. Here, we review the literature to discuss the potential usefulness of selected miRNAs as diagnostic and prognostic biomarkers in CRC.

Download Full-text

The OXL format for the exchange of integrated datasets

Journal of Integrative Bioinformatics ◽

10.1515/jib-2007-62 ◽

2007 ◽

Vol 4 (3) ◽

pp. 27-40 ◽

Cited By ~ 5

Author(s):

Jan Taubert ◽

Klaus Peter Sieren ◽

Matthew Hindle ◽

Berend Hoekman ◽

Rainer Winnenburg ◽

...

Keyword(s):

Life Science ◽

Original Data ◽

Biological Information ◽

Data Sets ◽

Complex Data ◽

Scientific Publications ◽

Mining System ◽

Biological Domain ◽

Data Source ◽

Text Mining System

Abstract A prerequisite for systems biology is the integration and analysis of heterogeneous experimental data stored in hundreds of life-science databases and millions of scientific publications. Several standardised formats for the exchange of specific kinds of biological information exist. Such exchange languages facilitate the integration process; however they are not designed to transport integrated datasets. A format for exchanging integrated datasets needs to i) cover data from a broad range of application domains, ii) be flexible and extensible to combine many different complex data structures, iii) include metadata and semantic definitions, iv) include inferred information, v) identify the original data source for integrated entities and vi) transport large integrated datasets. Unfortunately, none of the exchange formats from the biological domain (e.g. BioPAX, MAGE-ML, PSI-MI, SBML) or the generic approaches (RDF, OWL) fulfil these requirements in a systematic way.We present OXL, a format for the exchange of integrated data sets, and detail how the aforementioned requirements are met within the OXL format. OXL is the native format within the data integration and text mining system ONDEX. Although OXL was developed with the ONDEX system in mind, it also has the potential to be used in several other biological and non-biological applications described in this paper.Availability: The OXL format is an integral part of the ONDEX system which is freely available under the GPL at http://ondex.sourceforge.net/. Sample files can be found at http://prdownloads.sourceforge.net/ondex/ and the XML Schema at http://ondex.svn.sf.net/viewvc/*checkout*/ondex/trunk/backend/data/xml/ondex.xsd.

Download Full-text

Identification of Biomarkers Related To The Diagnosis And Prognosis of Thyroid Cancer Through Bioinformatics Analysis

10.21203/rs.3.rs-740573/v1 ◽

2021 ◽

Author(s):

Xiao-Li Xie ◽

Hua-Li Yin ◽

Yu-Lin Pan ◽

Guo-Xia Li ◽

Chun-Yan Yuan ◽

...

Keyword(s):

Gene Expression ◽

Thyroid Cancer ◽

Bioinformatics Analysis ◽

Prognostic Biomarkers ◽

Cancer Tissue ◽

Data Sets ◽

Hub Genes ◽

Normal Thyroid ◽

Key Genes ◽

Cancer Tissues

Abstract Background: Thyroid cancer is the most common malignant tumor of the head and neck. In recent years, the incidence of thyroid cancer (THCA) worldwide has rapidly increased and shows a trend in the younger generation. This study attempted to screen key genes and potential prognostic biomarkers for thyroid cancer using bioinformatics analysis.Methods: This study attempted to screen key genes and potential prognostic biomarkers for thyroid cancer using bioinformatics analysis. 101 cases of thyroid cancer and 78 cases of normal thyroid tissue were collected from three Gene Expression Omnibus (GEO) databases, then we identified the differentially expressed genes (DEGs) and conducted downstream analyses. Moreover, we screened hub genes by constructing a protein‐protein interaction (PPI) network. Finally, we assessed the expression level of hub genes in thyroid cancer tissue and its normal tissue using GEPIA and qRT-PCR respectively. Results: 159 upregulated and 251 downregulated genes were determined after gene integration of these three GEO data sets. Through PPI analysis, we consider the top 20 DEGs with high connectivity as the hub genes of THCA. After that, this study verified 20 central genes through the GEPIA database and found that only four hub genes (TOP2A, FN1, TIMP1, and MMP9) had significantly higher expression levels in thyroid cancer tissues than in normal thyroid tissues. We further analyzed the correlation between these four hub genes and the prognosis of patients with thyroid cancer, which suggests that FN1, MMP9, TIMP1 help assess the prognosis of patients with thyroid cancer. We performed GSEA analysis on these 4 hub genes simultaneously, found that the high expression of these 4 hub genes enriched the "cell cycle." Subsequently, we collected thyroid cancer tissue specimens, verified these four hub gene expression levels by RT-PCR, and found that only FN1 and TIMP1 genes in thyroid cancer tissues had significantly higher mRNA levels than normal tissues. Conclusions: Our research has identified 20 hub genes that may be related to the occurrence and development of thyroid cancer through multiple gene expression profile data sets and a series of comprehensive bioinformatics analyses. Further database and tissue validation analysis revealed that only 2 hub genes may be considered as potential prognostic biomarkers, including FN1 and TIMP1. In addition, these two hub genes are involved in the cell cycle, suggesting that they may play a role in the occurrence and development of thyroid cancer.

Download Full-text

Normalisation of Multicondition cDNA Macroarray Data

Comparative and Functional Genomics ◽

10.1155/2007/90578 ◽

2007 ◽

Vol 2007 ◽

pp. 1-12 ◽

Cited By ~ 3

Author(s):

Nicola L. Dawes ◽

Jarka Glassey

Keyword(s):

Biological Information ◽

Data Sets ◽

Dna Array ◽

Array Data ◽

Cdna Macroarray ◽

Regulatory Interactions ◽

Experimental Variation ◽

Single Colour ◽

Global Identification ◽

Normalisation Method

Background. Normalisation is a critical step in obtaining meaningful information from the high-dimensional DNA array data. This is particularly important when complex biological hypotheses/questions, such a functional analysis and regulatory interactions within biological systems, are investigated. A nonparametric, intensity-dependent normalisation method based on global identification of self-consistent set (SCS) of genes is proposed here for such systems.Results. The SCS normalisation is introduced and its behaviour demonstrated for a range of user-defined parameters affecting sits performance. It is compared to a standard global normalisation method in terms of noise reduction and signal retention.Conclusions. The SCS normalisation results using 16 macroarray data sets from aBacillus subtilisexperiment confirm that the method is capable of reducing undesirable experimental variation whilst retaining important biological information. The ease and speed of implementation mean that this method can be easily adapted to other multicondition time/strain series single colour array data.

Download Full-text