Classifying Lung Adenocarcinoma and Squamous Cell Carcinoma using RNA-Seq Data

Abstract

Background: Lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) are two primary subtypes of non-small cell lung carcinoma (NSCLC). Currently, the most widely used method to discriminate between LUAD and LUSC is hematoxylin-eosin (HE) staining. However, this method sometimes is unable to make the precise diagnosis on LUAD or LUSC. More accurate diagnostic approaches are highly desired. Methods: We propose to use gene expression profile to discriminate NSCLC patient's subtype. We leveraged RNA-Seq data from The Cancer Genome Atlas (TCGA) and randomly split the data into training and testing subsets. To construct classifiers based on the training data, we considered three methods: logistic regression on principal components (PCR), logistic regression with LASSO shrinkage (LASSO), and kth nearest neighbors (KNN). Performances of classifiers were evaluated and compared based on the testing data. Results: All gene expression-based classifiers show high accuracy in discriminating LUSC and LUAD. The classifier obtained by LASSO has the smallest overall misclassification rate of 3.42% (95% CI: 3.25%-3.60%) when using 0.5 as the cutoff value for the predicted probability of belonging to a subtype, followed by classifiers obtained by PCR (4.36%, 95% CI: 4.23%-4.49%) and KNN (8.70%, 95% CI: 8.57%-8.83%). The LASSO classifier also has the highest average area under the receiver operating characteristic curve (AUC) value of 0.993, compared to PCR (0.987) and KNN (0.965). Conclusions: Our results suggest that mRNA expressions are highly informative for classifying NSCLC subtypes and may potentially be used to assist clinical diagnosis.

LUAD And LUSC Cancer Classification, Biomarker Identification, and Pathway Analysis Using Overlapping Feature Selection Methods

“…Previous studies have utilized traditional feature selection and machine learning methods for cancer diagnosis, detection, and classi cation [10,11,19], but few have extended them to study potential biomarkers and biological pathways to discriminate between LUAD and LUSC. To improve cancer classi cation accuracy, novel machine learning, and feature selection methods have been developed [12,[20][21][22].…”

“…Multiple gene expression and immunohistochemistry studies have identi ed biological pathways and biomarkers that differentiate between LUAD and LUSC [6][7][8]. Other studies classi ed cancers using both novel and traditional machine learning or feature selection methods [9][10][11][12]. However, few have investigated cancers by applying multiple feature selection methods and selecting the overlapping features.…”

A survey of dimension reduction and classification methods for RNA-Seq data on malaria vector

“…In 2017, Zhengyan, Li, and Chi, [37] classified lung adenocarcinoma and Squamous cell carcinoma using RNA-Seq Data. They used gene expression profile to discriminate NSCLC (Non-Small Cell Lung Cancer) patient's subtype.…”

An 8-Gene Signature for Classifying Major Subtypes of Non-Small-Cell Lung Cancer

“…Using a method based on the nearest shrunken centroid classification algorithm, Charkiewicz et al 8 discovered a 53-gene signature for LUAD/LUSC classification and tested it on a single dataset. Huang et al 9 employed 3 different methods and reported high classification accuracies for the 3 methods. Employing a machine learning algorithm and using microarray data for training, Wu et al 10 discovered a 5-gene signature and tested their method/signature on RNA-Seq data from The Cancer Genome Atlas (TCGA).…”