Dataset Information

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.

SUBMITTER: Lee K

PROVIDER: S-EPMC6858312 | biostudies-literature | 2019 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

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

Lee Kanggeun K Jeong Hyoung-Oh HO Lee Semin S Jeong Won-Ki WK

Scientific reports 20191115 1

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, m ...[more]

PMID: 31729414

Similar Datasets

Project description:Development of machine learning solutions for prediction of functional and clinical significance of cancer driver genes and mutations are paramount in modern biomedical research and have gained a significant momentum in a recent decade. In this work, we integrate different machine learning approaches, including tree based methods, random forest and gradient boosted tree (GBT) classifiers along with deep convolutional neural networks (CNN) for prediction of cancer driver mutations in the genomic datasets. The feasibility of CNN in using raw nucleotide sequences for classification of cancer driver mutations was initially explored by employing label encoding, one hot encoding, and embedding to preprocess the DNA information. These classifiers were benchmarked against their tree-based alternatives in order to evaluate the performance on a relative scale. We then integrated DNA-based scores generated by CNN with various categories of conservational, evolutionary and functional features into a generalized random forest classifier. The results of this study have demonstrated that CNN can learn high level features from genomic information that are complementary to the ensemble-based predictors often employed for classification of cancer mutations. By combining deep learning-generated score with only two main ensemble-based functional features, we can achieve a superior performance of various machine learning classifiers. Our findings have also suggested that synergy of nucleotide-based deep learning scores and integrated metrics derived from protein sequence conservation scores can allow for robust classification of cancer driver mutations with a limited number of highly informative features. Machine learning predictions are leveraged in molecular simulations, protein stability, and network-based analysis of cancer mutations in the protein kinase genes to obtain insights about molecular signatures of driver mutations and enhance the interpretability of cancer-specific classification models.

Project description:Osteosarcoma accounts for 28% of primary bone malignancies in adults and up to 56% in children and adolescents (<20 years). However, early diagnosis and treatment are still inadequate, and new improvements are still needed. Missed diagnoses exist due to fewer traditional diagnostic methods, and clinical symptoms are often already present before diagnosis. This study aimed to develop novel and efficient predictive models for the diagnosis of osteosarcoma and to identify potential targets for exploring osteosarcoma markers. First, osteosarcoma and normal tissue expression microarray datasets were downloaded from the Gene Expression Omnibus (GEO). Then we screened the differentially expressed genes (DEGs) in the osteosarcoma and normal groups in the training group. Next, in order to explore the biologically relevant role of DEGs, Metascape and enrichment analyses were also performed on DEGs. The "randomForest" and "neuralnet" packages in R software were used to select representative genes and construct diagnostic models for osteosarcoma. The next step is to validate the model of the artificial neural network. Then, we performed an immune infiltration analysis by using the training set data. Finally, we constructed a prognostic model using representative genes for prognostic analysis. The copy number of osteosarcoma was also analyzed. A random forest classifier identified nine representative genes (ANK1, TGFBR3, TNFRSF21, HSPB8, ITGA7, RHD, AASS, GREM2, NFASC). HSPB8, RHD, AASS, and NFASC were genes we identified that have not been previously reported to be associated with osteosarcoma. The osteosarcoma diagnostic model we constructed has good performance with areas under the curves (AUCs) of 1 and 0.987 in the training and validation groups, respectively. This study opens new horizons for the early diagnosis of osteosarcoma and provides representative markers for the future treatment of osteosarcoma. This is the first study to pioneer the establishment of a genetic diagnosis model for osteosarcoma and advance the development of osteosarcoma diagnosis and treatment.

Dataset Information

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

Publications

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

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets