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.
Highlights
With recent advances in DNA sequencing technologies, fast acquisition of large-scale genomic data has become commonplace
We propose a novel cancer type classification method, the Cancer Predictor using an Ensemble Model (CPEM), which is based on the combination of advanced machine learning algorithms with various types of cancer somatic alterations and their derived features as input
To identify which input features are useful in classification, we conducted a comprehensive study on how various genetic alterations, such as gene-level mutation profiles, mutation rates, mutation spectra and signatures, and gene-level copy number alterations affect the accuracy of the classifier and we discuss the biological reasoning behind this result
Summary
With recent advances in DNA sequencing technologies, fast acquisition of large-scale genomic data has become commonplace. 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. Using the tissue-specific nature of somatic alterations in cancer, a number of prediction methods for cancer type were recently developed by employing machine learning classifiers and various mutation features to improve classification accuracy. Most of them were able to handle only a small subset of the cancer types from the database for classification To address these problems, we propose a novel cancer type classification method, the Cancer Predictor using an Ensemble Model (CPEM) (see Fig. 1), which is based on the combination of advanced machine learning algorithms with various types of cancer somatic alterations and their derived features as input. To the best of our knowledge, the proposed method is the most accurate multiclass cancer type classification method yet-devised that is able to predict the largest class of cancer types, covering the entire cancer list in the TCGA database[3]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
