Comprehensive and critical evaluation of individualized pathway activity measurement tools on pan-cancer data

Abstract The identification of hidden responders is often an essential challenge in precision oncology. A recent attempt based on machine learning has been proposed for classifying aberrant pathway activity from multiomic cancer data. However, we note several critical limitations there, such as high-dimensionality, data sparsity and model performance. Given the central importance and broad impact of precision oncology, we propose nature-inspired deep Ras activation pan-cancer (NatDRAP), a deep neural network (DNN) model, to address those restrictions for the identification of hidden responders. In this study, we develop the nature-inspired deep learning model that integrates bulk RNA sequencing, copy number and mutation data from PanCanAltas to detect pan-cancer Ras pathway activation. In NatDRAP, we propose to synergize the nature-inspired artificial bee colony algorithm with different gradient-based optimizers in one framework for optimizing DNNs in a collaborative manner. Multiple experiments were conducted on 33 different cancer types across PanCanAtlas. The experimental results demonstrate that the proposed NatDRAP can provide superior performance over other benchmark methods with strong robustness towards diagnosing RAS aberrant pathway activity across different cancer types. In addition, gene ontology enrichment and pathological analysis are conducted to reveal novel insights into the RAS aberrant pathway activity identification and characterization. NatDRAP is written in Python and available at https://github.com/lixt314/NatDRAP1.

Download Full-text

A novel unsupervised learning model for detecting driver genes from pan-cancer data through matrix tri-factorization framework with pairwise similarities constraints

Neurocomputing ◽

10.1016/j.neucom.2018.03.026 ◽

2018 ◽

Vol 296 ◽

pp. 64-73 ◽

Cited By ~ 7

Author(s):

Jianing Xi ◽

Ao Li ◽

Minghui Wang

Keyword(s):

Unsupervised Learning ◽

Learning Model ◽

Driver Genes ◽

Cancer Data ◽

Pan Cancer

Download Full-text

The Biological Function Delineated Across Pan-Cancer Levels Through lncRNA-Based Prognostic Risk Assessment Factors for Pancreatic Cancer

Frontiers in Cell and Developmental Biology ◽

10.3389/fcell.2021.694652 ◽

2021 ◽

Vol 9 ◽

Author(s):

Xudong Tang ◽

Mengyan Zhang ◽

Liang Sun ◽

Fengyan Xu ◽

Xin Peng ◽

...

Keyword(s):

Pancreatic Cancer ◽

Prognostic Marker ◽

Cancer Genome ◽

The Cancer Genome Atlas ◽

Research Network ◽

Molecular Characteristics ◽

Cancer Data ◽

Cancer Genome Atlas ◽

Pan Cancer ◽

Genome Atlas

Long non-coding RNAs (lncRNAs) play key roles in tumors and function not only as important molecular markers for cancer prognosis, but also as molecular characteristics at the pan-cancer level. Because of the poor prognosis of pancreatic cancer, accurate assessment of prognosis is a key issue in the development of treatment plans for pancreatic cancer. Here we analyzed pancreatic cancer data from The Cancer Genome Atlas and The Genotype Tissue Expression database using Cox regression and lasso regression in analyses using a combination of the two databases as well as only The Cancer Genome Atlas database (Cancer Genome Atlas Research Network et al., 2013). A prognostic risk score model with significant correlation with pancreatic cancer survival was constructed, and two lncRNAs were investigated. Additional analysis of 33 cancers using the two lncRNAs showed that lncRNA TsPOAP1-AS1 was a prognostic marker of seven cancers, among which pancreatic cancer was the most significant, and lncRNA mi600hg was a prognostic marker of ovarian cancer and pancreatic cancer. LncRNA TsPOAP1-AS1 is associated with clinical stage and tumor mutation burden of some cancers as well as a strong degree of immune infiltration in many cancers, while a strong correlation between lncRNA mi600hg and microsatellite instability was observed in several cancers. The results of this study help further our understanding of the different functions of lncRNAs in cancer and may aid in the clinical application of lncRNAs as prognostic factors for cancer.

Download Full-text

The Cancer Genomic Atlas – “TO CONQUER CANCER”

International Journal of Molecular and Immuno Oncology ◽

10.25259/ijmio_28_2020 ◽

2020 ◽

Vol 0 ◽

pp. 1-6

Author(s):

Sai Sri Kavya Kadali ◽

Rachna Gowlikar ◽

Syeda Nooreen Fatima

Keyword(s):

Genetic Basis ◽

Data Repository ◽

Rare Cancer ◽

Cancer Data ◽

Data Collection Process ◽

Genomics And Proteomics ◽

Prevention Studies ◽

Cancer Types ◽

Pan Cancer

The Cancer Genomic Atlas (TCGA) is a publicly accessible cancer data repository and tool that allows us to understand the molecular basis of cancer through the application of genomics and proteomics. So far, researchers have been able to diagnose 33 cancer types including 10 rare cancer types. The key features of TCGA are to make the data collection process publicly accessible for the better understanding of the molecular and genetic basis of cancer and its mechanism of action along with its prevention. Studies on different cancer types along with comprehensive pan cancer analysis have expanded the understanding and purpose of TCGA. Ever since its’ conceptualization, its’ high-throughput approach has provided a platform for the identification of genes and pathways involved in cancers and accurate classification of cancers.

Download Full-text

Comprehensive Analysis to Identify SPP1 as a Prognostic Biomarker in Cervical Cancer

Frontiers in Genetics ◽

10.3389/fgene.2021.732822 ◽

2022 ◽

Vol 12 ◽

Author(s):

Kaidi Zhao ◽

Zhou Ma ◽

Wei Zhang

Keyword(s):

Cervical Cancer ◽

Roc Curve ◽

Immune Cell ◽

Cox Regression ◽

Prognostic Biomarker ◽

R Package ◽

Cancer Data ◽

Immune Infiltration ◽

Pan Cancer ◽

Geo Database

Background:SPP1, secreted phosphoprotein 1, is a member of the small integrin-binding ligand N-linked glycoprotein (SIBLING) family. Previous studies have proven SPP1 overexpressed in a variety of cancers and can be identified as a prognostic factor, while no study has explored the function and carcinogenic mechanism of SPP1 in cervical cancer.Methods: We aimed to demonstrate the relationship between SPP1 expression and pan-cancer using The Cancer Genome Atlas (TCGA) database. Next, we validated SPP1 expression of cervical cancer in the Gene Expression Omnibus (GEO) database, including GSE7803, GSE63514, and GSE9750. The receiver operating characteristic (ROC) curve was used to evaluate the feasibility of SPP1 as a differentiating factor by the area under curve (AUC) score. Cox regression and logistic regression were performed to evaluate factors associated with prognosis. The SPP1-binding protein network was built by the STRING tool. Enrichment analysis by the R package clusterProfiler was used to explore potential function of SPP1. The single-sample GSEA (ssGSEA) method from the R package GSVA and TIMER database were used to investigate the association between the immune infiltration level and SPP1 expression in cervical cancer.Results: Pan-cancer data analysis showed that SPP1 expression was higher in most cancer types, including cervical cancer, and we got the same result in the GEO database. The ROC curve suggested that SPP1 could be a potential diagnostic biomarker (AUC = 0.877). High SPP1 expression was associated with poorer overall survival (OS) (P = 0.032). Further enrichment and immune infiltration analysis revealed that high SPP1 expression was correlated with regulating the infiltration level of neutrophil cells and some immune cell types, including macrophage and DC.Conclusion:SPP1 expression was higher in cervical cancer tissues than in normal cervical epithelial tissues. It was significantly associated with poor prognosis and immune cell infiltration. Thus, SPP1 may become a promising prognostic biomarker for cervical cancer patients.

Download Full-text