ExampleRawData: Example dataset to test the performance of the sigFeature...

In sigFeature: sigFeature: Significant feature selection using SVM-RFE & t-statistic

Description

For this significant feature selection procedure, microarray data (GSE2280) from patients with squamous cell carcinoma of the oral cavity (OSCC) has been used( O'Donnell RK et al. (2005)). Affymetrix Human Genome Array, U133A was selected for genome-wide transcription analysis for this data set. In this paper, the gene expression profiles obtained from primary squamous cell carcinoma of the oral cavity (OSCC) that were metastatic to lymph nodes (N+) compared to those that were not metastatic (N-). A total of 18 OSCCs were analyzed for gene expression. In their analysis a predictive rule was built using a support vector machine, and the accuracy of the rule was evaluated using cross-validation the original data set and prediction of an independent set of four patients. A signature gene set is produced which is able to predict the four independent patients correctly as well as associating five lymph node metastases from the original patient set with the metastatic primary tumour group.

Usage

data("ExampleRawData")

Format

    The format is:
    num [1:27, 1:2205] 72.5 177.2 75.7 128.9 142 ...
    - attr(*, "dimnames")=List of 2
    ..$ : chr [1:27] "GSM42246" "GSM42248" "GSM42250" "GSM42252" ...
    ..$ : chr [1:2205] "1494_f_at" "179_at" "200014_s_at" "200059_s_at" ...

Details

For "sigFeature" package evaluation, the microarray dataset has been classified into two classes such as lymph node metastatic (N+) and No lymph node metastatic (N-) (according to the TNM staging), provided in the dataset. After downloading the data set from GEO database firstly, it was normalized using the "quantile" normalization method using the Bioconductor package "limma". To reduce the runtime of this sigFeature function, a subset of the total dataset is taken by the ratio between the difference between two groups with cut off value (p-value 0.07). Now the expression value of the sub-dataset is considered here as "x". The patients without lymph node metastasis are represented -1, and the patients with lymph node metastasis are represented as 1. Those -1 and 1 value is incorporated with "y" as sample labels.

Value

ExampleRawData

Return the values stored in the variable.

Source

https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE2280

References

O'Donnell RK, Kupferman M, Wei SJ, Singhal S et al. Gene expression signature predicts lymphatic metastasis in squamous cell carcinoma of the oral cavity. Oncogene 2005 Feb 10;24(7):1244-51. PMID: 15558013

Examples

data("ExampleRawData")

sigFeature documentation built on Nov. 8, 2020, 8:28 p.m.