Pathway-based biomarker identification with crosstalk analysis for robust prognosis prediction in hepatocellular carcinoma

Botao Fa, Chengwen Luo, Zhou Tang, Yuting Yan, Yue Zhang, Zhangsheng Yu

Research output: Contribution to journalArticle

Abstract

Background: Although many prognostic single-gene (SG) lists have been identified in cancer research, application of these features is hampered due to poor robustness and performance on independent datasets. Pathway-based approaches have thus emerged which embed biological knowledge to yield reproducible features. Methods: Pathifier estimates pathways deregulation score (PDS) to represent the extent of pathway deregulation based on expression data, and most of its applications treat pathways as independent without addressing the effect of gene overlap between pathway pairs which we refer to as crosstalk. Here, we propose a novel procedure based on Pathifier methodology, which for the first time has been utilized with crosstalk accommodated to identify disease-specific features to predict prognosis in patients with hepatocellular carcinoma (HCC). Findings: With the cohort (N = 355) of HCC patients from The Cancer Genome Atlas (TCGA), cross validation (CV) revealed that PDSs identified were more robust and accurate than the SG features by deep learning (DL)-based approach. When validated on external HCC datasets, these features outperformed the SGs consistently. Interpretation: On average, we provide 10.2% improvement of prediction accuracy. Importantly, governing genes in these features provide valuable insight into the cancer hallmarks of HCC. We develop an R package PATHcrosstalk (available from GitHub https://github.com/fabotao/PATHcrosstalk) with which users can discover pathways of interest with crosstalk effect considered.

Original languageSpanish
JournalEBioMedicine
DOIs
StatePublished - Jan 1 2019
Externally publishedYes

Fingerprint

Biomarkers
Crosstalk
Hepatocellular Carcinoma
Genes
Deregulation
Neoplasms
Atlases
Learning
Genome
Deep learning
Research
Datasets

Keywords

  • Biomarker
  • Crosstalk
  • Deep learning
  • Hepatocellular carcinoma
  • Overall survival
  • Pathway-based

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology(all)

Cite this

Pathway-based biomarker identification with crosstalk analysis for robust prognosis prediction in hepatocellular carcinoma. / Fa, Botao; Luo, Chengwen; Tang, Zhou; Yan, Yuting; Zhang, Yue; Yu, Zhangsheng.

In: EBioMedicine, 01.01.2019.

Research output: Contribution to journalArticle

@article{31a3d91dcbb543a5b3e67deb8852a4d2,
title = "Pathway-based biomarker identification with crosstalk analysis for robust prognosis prediction in hepatocellular carcinoma",
abstract = "Background: Although many prognostic single-gene (SG) lists have been identified in cancer research, application of these features is hampered due to poor robustness and performance on independent datasets. Pathway-based approaches have thus emerged which embed biological knowledge to yield reproducible features. Methods: Pathifier estimates pathways deregulation score (PDS) to represent the extent of pathway deregulation based on expression data, and most of its applications treat pathways as independent without addressing the effect of gene overlap between pathway pairs which we refer to as crosstalk. Here, we propose a novel procedure based on Pathifier methodology, which for the first time has been utilized with crosstalk accommodated to identify disease-specific features to predict prognosis in patients with hepatocellular carcinoma (HCC). Findings: With the cohort (N = 355) of HCC patients from The Cancer Genome Atlas (TCGA), cross validation (CV) revealed that PDSs identified were more robust and accurate than the SG features by deep learning (DL)-based approach. When validated on external HCC datasets, these features outperformed the SGs consistently. Interpretation: On average, we provide 10.2{\%} improvement of prediction accuracy. Importantly, governing genes in these features provide valuable insight into the cancer hallmarks of HCC. We develop an R package PATHcrosstalk (available from GitHub https://github.com/fabotao/PATHcrosstalk) with which users can discover pathways of interest with crosstalk effect considered.",
keywords = "Biomarker, Crosstalk, Deep learning, Hepatocellular carcinoma, Overall survival, Pathway-based",
author = "Botao Fa and Chengwen Luo and Zhou Tang and Yuting Yan and Yue Zhang and Zhangsheng Yu",
year = "2019",
month = "1",
day = "1",
doi = "10.1016/j.ebiom.2019.05.010",
language = "Spanish",
journal = "EBioMedicine",
issn = "2352-3964",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - Pathway-based biomarker identification with crosstalk analysis for robust prognosis prediction in hepatocellular carcinoma

AU - Fa, Botao

AU - Luo, Chengwen

AU - Tang, Zhou

AU - Yan, Yuting

AU - Zhang, Yue

AU - Yu, Zhangsheng

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Background: Although many prognostic single-gene (SG) lists have been identified in cancer research, application of these features is hampered due to poor robustness and performance on independent datasets. Pathway-based approaches have thus emerged which embed biological knowledge to yield reproducible features. Methods: Pathifier estimates pathways deregulation score (PDS) to represent the extent of pathway deregulation based on expression data, and most of its applications treat pathways as independent without addressing the effect of gene overlap between pathway pairs which we refer to as crosstalk. Here, we propose a novel procedure based on Pathifier methodology, which for the first time has been utilized with crosstalk accommodated to identify disease-specific features to predict prognosis in patients with hepatocellular carcinoma (HCC). Findings: With the cohort (N = 355) of HCC patients from The Cancer Genome Atlas (TCGA), cross validation (CV) revealed that PDSs identified were more robust and accurate than the SG features by deep learning (DL)-based approach. When validated on external HCC datasets, these features outperformed the SGs consistently. Interpretation: On average, we provide 10.2% improvement of prediction accuracy. Importantly, governing genes in these features provide valuable insight into the cancer hallmarks of HCC. We develop an R package PATHcrosstalk (available from GitHub https://github.com/fabotao/PATHcrosstalk) with which users can discover pathways of interest with crosstalk effect considered.

AB - Background: Although many prognostic single-gene (SG) lists have been identified in cancer research, application of these features is hampered due to poor robustness and performance on independent datasets. Pathway-based approaches have thus emerged which embed biological knowledge to yield reproducible features. Methods: Pathifier estimates pathways deregulation score (PDS) to represent the extent of pathway deregulation based on expression data, and most of its applications treat pathways as independent without addressing the effect of gene overlap between pathway pairs which we refer to as crosstalk. Here, we propose a novel procedure based on Pathifier methodology, which for the first time has been utilized with crosstalk accommodated to identify disease-specific features to predict prognosis in patients with hepatocellular carcinoma (HCC). Findings: With the cohort (N = 355) of HCC patients from The Cancer Genome Atlas (TCGA), cross validation (CV) revealed that PDSs identified were more robust and accurate than the SG features by deep learning (DL)-based approach. When validated on external HCC datasets, these features outperformed the SGs consistently. Interpretation: On average, we provide 10.2% improvement of prediction accuracy. Importantly, governing genes in these features provide valuable insight into the cancer hallmarks of HCC. We develop an R package PATHcrosstalk (available from GitHub https://github.com/fabotao/PATHcrosstalk) with which users can discover pathways of interest with crosstalk effect considered.

KW - Biomarker

KW - Crosstalk

KW - Deep learning

KW - Hepatocellular carcinoma

KW - Overall survival

KW - Pathway-based

UR - http://www.scopus.com/inward/record.url?scp=85065516732&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85065516732&partnerID=8YFLogxK

U2 - 10.1016/j.ebiom.2019.05.010

DO - 10.1016/j.ebiom.2019.05.010

M3 - Article

C2 - 31101593

AN - SCOPUS:85065516732

JO - EBioMedicine

JF - EBioMedicine

SN - 2352-3964

ER -