2020
DOI: 10.1042/bsr20201482
|View full text |Cite
|
Sign up to set email alerts
|

Survival prediction in patients with colon adenocarcinoma via multiomics data integration using a deep learning algorithm

Abstract: This study proposed a deep learning (DL) algorithm to predict survival in patients with colon adenocarcinoma (COAD) based on multi-omics integration. The survival-sensitive model was constructed using an autoencoder for DL implementation based on The Cancer Genome Atlas (TCGA) data of patients with COAD. The autoencoder framework was compared to PCA, NMF, t-SNE, and univariable Cox-PH model for identifying survival-related features. The prognostic robustness of the inferred survival risk groups was validated u… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
36
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
5
2
1

Relationship

0
8

Authors

Journals

citations
Cited by 26 publications
(36 citation statements)
references
References 37 publications
0
36
0
Order By: Relevance
“…Subsequently, in the five studies (44, 49, 58, 65, 76) that integrated the information of multi-omics biomarkers including DNA methylation, the selected prognosis-associated hidden nodes were used to cluster samples by K-means clustering algorithm. After that, top 20 to 100 significantly differently methylated genes were identified by comparing across clusters.…”
Section: Resultsmentioning
confidence: 99%
See 1 more Smart Citation
“…Subsequently, in the five studies (44, 49, 58, 65, 76) that integrated the information of multi-omics biomarkers including DNA methylation, the selected prognosis-associated hidden nodes were used to cluster samples by K-means clustering algorithm. After that, top 20 to 100 significantly differently methylated genes were identified by comparing across clusters.…”
Section: Resultsmentioning
confidence: 99%
“…We recognize that similar ML approaches can be applied to non-genome-wide studies or to the identification of other biomarkers (e.g., mRNA) relevant to other diseases. Second, in the included multi-omics studies (44, 49, 58, 65, 76), identifying individual prognostically relevant DNA methylation biomarkers were merely side products instead of primary study aims. Third, variations in disease outcomes and the lack of benchmarking studies prevented us from comparing the performance of identified workflows.…”
Section: Limitation Of the Studymentioning
confidence: 99%
“…Very few data are available on the involvement of GPM6A in cancers. While GPM6A has been identified as a potential oncogene in lymphoid leukemia [ 12 ], and contributes to the poor prognosis of colorectal cancer [ 14 , 15 ], its role in GB has never been reported. First, we showed that GPM6A is overexpressed in the invasive GBSC compared to non-invasive cells and localized in lamellipodia/pseudopodia-like structures, suggesting a role in cell migration/invasion.…”
Section: Discussionmentioning
confidence: 99%
“…In colorectal cancer, the up-regulation of GPM6A was closely related to a poorer overall survival. In addition, a higher expression of GPM6A was observed in the poorly differentiated compared to the highly differentiated colorectal carcinoma tissues [ 14 , 15 ].…”
Section: Introductionmentioning
confidence: 99%
“…Eight compounds with low nanomolar IC 50 Glycoprotein M 6 a (Gpm 6 a) is a neuronal surface protein involved in differentiation and migration of neuronal stem cells, giving it roles in neuronal plasticity, neurite and filopodia outgrowth and motility, and probably also in synapse formation [12][13][14]. The peer reviewed literature contains little about the involvement of Gpm 6 a in cancer beyond its overexpression in some carcinomas [15,16] and a possible oncogenic role in lymphoid leukemia [17] Wdr5 is a highly-conserved nuclear protein that performs multiple scaffolding functions in the context of chromatin and forms a catalytically active core complex with Mll, RbBP5 and Ash2L, each of which is a common component of all known human histone H3-K4 methylating complexes [18]. Knockdown of Wdr5 results in a significant decrease in the levels of H3-K4 trimethylation.…”
mentioning
confidence: 99%