Modeling Gene Regulation in Liver Hepatocellular Carcinoma with Random Forests

Kazan, Hilal

doi:10.1155/2016/1035945

Cited by 7 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This further supports the notion that the deregulation of the CRL2 pVHL complex in HCC is not due to a single common cause, but rather several causes that can compound to affect patient outcomes. The multimodal nature of this regulation has also been observed in HCC more generally, where models that incorporate data on CN variations, hypomethylation, and miRNA expression more accurately predict mRNA expression than those that discard any individual mode ( Kazan, 2016 ). Similar forms of multimodal deregulation have also been observed in oncogenes outside of the HCC context, including EGFR , MDM2 , and PDGFRA ( Louhimo and Hautaniemi, 2011 ), as well as other UPP genes ( Ge et al, 2018 ).…”

Section: Discussionmentioning

confidence: 92%

Genetic and Epigenetic Mechanisms Deregulate the CRL2pVHL Complex in Hepatocellular Carcinoma

Minatel¹,

Cohn²,

Pewarchuk³

et al. 2022

Front. Genet.

View full text Add to dashboard Cite

Dysregulation of ubiquitin-proteasome pathway genes through copy number alteration, promoter hypomethylation, and miRNA deregulation is involved in cancer development and progression. Further characterizing alterations in these genes may uncover novel drug targets across a range of diseases in which druggable alterations are uncommon, including hepatocellular carcinoma (HCC). We analyzed 377 HCC and 59 adjacent non-malignant liver tissue samples, focusing on alterations to component genes of the widely studied CRL2pVHL E3 ubiquitin ligase complex. mRNA upregulation of the component genes was common, and was correlated with DNA hypomethylation and copy number increase, but many tumours displayed overexpression that was not explained by either mechanism. Interestingly, we found 66 miRNAs, including 39 previously unannotated miRNAs, that were downregulated in HCC and predicted to target one or more CRL2pVHL components. Several miRNAs, including hsa-miR-101-3p and hsa-miR-139-5p, were negatively correlated with multiple component genes, suggesting that miRNA deregulation may contribute to CRL2pVHL overexpression. Combining miRNA and mRNA expression, DNA copy number, and methylation status into one multidimensional survival analysis, we found a significant association between greater numbers of alterations and poorer overall survival for multiple component genes. While the intricacies of CRL2pVHL complex gene regulation require additional research, it is evident that multiple causes for the deregulation of these genes must be considered in HCC, including non-traditional mechanisms.

show abstract

Section: Discussionmentioning

confidence: 92%

Genetic and Epigenetic Mechanisms Deregulate the CRL2pVHL Complex in Hepatocellular Carcinoma

Minatel¹,

Cohn²,

Pewarchuk³

et al. 2022

Front. Genet.

View full text Add to dashboard Cite

show abstract

“…The random forest (RF) algorithm is an ensemble machine learning method developed by Breiman [ 24 ]. It has been widely applied to prediction problems in bioinformatics [ 25 – 27 ]. The RF algorithm consists of multibase tree-structured classifiers such as CART (classification and regression tree), and it is robust to noise, is not hindered by overfitting, and is computationally feasible.…”

Section: Methodsmentioning

confidence: 99%

Comparative Analysis and Classification of Cassette Exons and Constitutive Exons

Cui

Cai

Stanley

2017

BioMed Research International

View full text Add to dashboard Cite

Alternative splicing (AS) is a major engine that drives proteome diversity in mammalian genomes and is a widespread cause of human hereditary diseases. More than 95% of genes in the human genome are alternatively spliced, and the most common type of AS is the cassette exon. Recent discoveries have demonstrated that the cassette exon plays an important role in genetic diseases. To discover the formation mechanism of cassette exon events, we statistically analyze cassette exons and find that cassette exon events are strongly influenced by individual exons that are smaller in size and that have a lower GC content, more codon terminations, and weaker splice sites. We propose an improved random-forest-based hybrid method of distinguishing cassette exons from constitutive exons. Our method achieves a high accuracy in classifying cassette exons and constitutive exons and is verified to outperform previous approaches. It is anticipated that this study will facilitate a better understanding of the underlying mechanisms in cassette exons.

show abstract

“…the human gut microbiome and detection of cancers such as ovarian, lung and breast 22 , 23 . Although random forest (RF) has been useful in selection of genetic features for HCC detection 24 , to our knowledge, RF has not yet been used to build an HCC screening or diagnosis test using multiple biomarkers. Because of RF’s strength in biological applications of machine learning (e.g.…”

Section: Introductionmentioning

confidence: 99%

Development and Evaluation of Novel Statistical Methods in Urine Biomarker-Based Hepatocellular Carcinoma Screening

Wang

Jain

Chen³

et al. 2018

Sci Rep

View full text Add to dashboard Cite

Hepatocellular carcinoma is one of the fastest growing cancers in the US and has a low survival rate, partly due to difficulties in early detection. Because of HCC’s high heterogeneity, it has been suggested that multiple biomarkers would be needed to develop a sensitive HCC screening test. This study applied random forest (RF), a machine learning technique, and proposed two novel models, fixed sequential (FS) and two-step (TS), for comparison with two commonly used statistical techniques, logistic regression (LR) and classification and regression trees (CART), in combining multiple urine DNA biomarkers for HCC screening using biomarker values obtained from 137 HCC and 431 non-HCC (224 hepatitis and 207 cirrhosis) subjects. The sensitivity, specificity, area under the receiver operating curve, and variability were estimated through repeated 10-fold cross-validation to compare the models’ performances in accuracy and robustness. We show that RF and TS have higher accuracy and stability; specifically, they reach 90% specificity and 86%/87% sensitivity respectively along with 15% higher sensitivity and 10% higher specificity than LR in cross-validation. The potential of RF and TS to develop a panel of multiple biomarkers and the possibility for self-training, cloud-based models for HCC screening are discussed.

show abstract

Modeling Gene Regulation in Liver Hepatocellular Carcinoma with Random Forests

Cited by 7 publications

References 25 publications

Genetic and Epigenetic Mechanisms Deregulate the CRL2pVHL Complex in Hepatocellular Carcinoma

Genetic and Epigenetic Mechanisms Deregulate the CRL2pVHL Complex in Hepatocellular Carcinoma

Comparative Analysis and Classification of Cassette Exons and Constitutive Exons

Development and Evaluation of Novel Statistical Methods in Urine Biomarker-Based Hepatocellular Carcinoma Screening

Contact Info

Product

Resources

About