CPEM: Accurate cancer type classification based on somatic alterations using an ensemble of a random forest and a deep neural network

Abstract

With recent advances in DNA sequencing technologies, fast acquisition of large-scale genomic data has become commonplace. For cancer studies, in particular, there is an increasing need for the classification of cancer type based on somatic alterations detected from sequencing analyses. However, the ever-increasing size and complexity of the data make the classification task extremely challenging. In this study, we evaluate the contributions of various input features, such as mutation profiles, mutation rates, mutation spectra and signatures, and somatic copy number alterations that can be derived from genomic data, and further utilize them for accurate cancer type classification. We introduce a novel ensemble of machine learning classifiers, called CPEM (Cancer Predictor using an Ensemble Model), which is tested on 7,002 samples representing over 31 different cancer types collected from The Cancer Genome Atlas (TCGA) database. We first systematically examined the impact of the input features. Features known to be associated with specific cancers had relatively high importance in our initial prediction model. We further investigated various machine learning classifiers and feature selection methods to derive the ensemble-based cancer type prediction model achieving up to 84% classification accuracy in the nested 10-fold cross-validation. Finally, we narrowed down the target cancers to the six most common types and achieved up to 94% accuracy.

Cancer Classification with a Cost-Sensitive Naive Bayes Stacking Ensemble

“…Therefore, a large number of classification models based on ensemble learning have been proposed. For example, Lee et al [ 5 ] developed an ensemble model based on random forest and deep neural network for cancer classification and achieved an accuracy of 94%. ALzubi et al [ 28 ] used the boosted weighted optimization neural network ensemble classification algorithm to classify cancer patients, thereby improving the accuracy of cancer diagnosis.…”

“…At present, the application of machine learning methods to cancer classification is a significant research field in bioinformatics [4,5]. Many traditional machine learning methods have been successfully applied to the classification analysis of gene expression data [6][7][8][9], such as RF, decision tree, KNN, and neural networks.…”

Predicting chemotherapy response using a variational autoencoder approach

“…While in this work we focused on tumor transcriptome data which can be measured with high precision over a wide dynamic range of transcript abundances by RNA-seq, we note that TCGA datasets of tumor somatic mutations and copy number alteration events are also available (Hutter and Zenklusen, 2018). Given the voluminous literature on the use of tumor somatic genomic data for precision cancer diagnosis (Mitchel et al , 2019; Zhang et al , 2020; Lee et al , 2019), tumor DNA datasets are fertile ground for developing a semi-supervised, multi-omics model for predicting response to chemotherapy.…”

Mbl-2 gene polymorphisms in pediatric Burkitt lymphoma: an approach based on machine learning techniques

“…In the context of supervised analysis, some authors have been using the cross-validation strategy to classify some types of cancer, for example Lee et al, (2019), who used the RF with 10-fold cross-validation to classify 31 types of cancer and managed to reach 84% accuracy, reaching up to 94% for the 6 most common types of cancer. As mentioned, RF was used in our study because it avoids overfitting more efficiently than decision trees, in addition to obtaining greater accuracy and being more stable (J. C. da Silva, 2018).…”