Cancer type prediction based on copy number aberration and chromatin 3D structure with convolutional neural networks

Yuan, Yuhong; Shi, Yi; Su, Xianbin; Zou, Xin; Luo, Qing; Feng, David Dagan; Cai, Weidong; Han, Ze‐Guang

doi:10.1186/s12864-018-4919-z

Cited by 28 publications

(26 citation statements)

References 42 publications

(46 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We will continue our work to investigate this issue by introducing yet another rich source of transcriptomic data from GTEx collection [33]. Furthermore, as suggested by previous studies [13,15,[34][35][36][37], we may incorporate additional genome-wide profiling information, such as DNA mutation, copy number variation, and DNA methylation as additional input matrices to enrich the complexity for model training, and thus to improve the classification accuracy.…”

Section: Discussionmentioning

confidence: 99%

Convolutional neural network models for cancer type prediction based on gene expression

Chiu²,

et al. 2020

View full text Add to dashboard Cite

BackgroundPrecise prediction of cancer types is vital for cancer diagnosis and therapy. Important cancer marker genes can be inferred through predictive model. Several studies have attempted to build machine learning models for this task however none has taken into consideration the effects of tissue of origin that can potentially bias the identification of cancer markers. ResultsIn this paper, we introduced several Convolutional Neural Network (CNN) models that take unstructured gene expression inputs to classify tumor and non-tumor samples into their designated cancer types or as normal. Based on different designs of gene embeddings and convolution schemes, we implemented three CNN models: 1D-CNN, 2D-Vanilla-CNN, and 2D-Hybrid-CNN. The models were trained and tested on combined 10,340 samples of 33 cancer types and 731 matched normal tissues of The Cancer Genome Atlas (TCGA). Our models achieved excellent prediction accuracies (93.9-95.0%) among 34 classes (33 cancers and normal). Furthermore, we interpreted one of the models, known as 1D-CNN model, with a guided saliency technique and identified a total of 2,090 cancer markers (108 per class). The concordance of differential expression of these markers between the cancer type they represent and others is confirmed. In breast cancer, for instance, our model identified well-known markers, such as GATA3 and ESR1. Finally, we extended the 1D-CNN model for prediction of breast cancer subtypes and achieved an average accuracy of 88.42% among 5 subtypes. The codes can be found at -3 -https://github.com/chenlabgccri/CancerTypePrediction. ConclusionsHere we present novel CNN designs for accurate and simultaneous cancer/normal and cancer types prediction based on gene expression profiles, and unique model interpretation scheme to elucidate biologically relevance of cancer marker genes after eliminating the effects of tissue-of-origin. The proposed model had light hyperparameters to be trained and thus can be easily adapt to facilitate cancer diagnosis in the future. BackgroundCancer is the second leading cause of death worldwide, an average of one in six deaths is due to cancer [1]. Considerable research efforts have been devoted to cancer diagnosis and treatment techniques to lessen its impact on human health. Cancer prediction's major focus is on cancer susceptibility, recurrence, and prognosis, while the aim of cancer detection is the classification of tumor types and identification of markers for each cancer such that we can build a learning machine to identify specific metastatic tumor type or detect cancer at their earlier stage. With the increased awareness of precision medicine and early detection techniques matured over years of technology development [2][3][4], including particularly many detection screens achieving a sensitivity around 70-80% [5], the demand for applying novel machine learning methods to discover new biomarkers has become one of the key driving factors in many clinical and translational applications.Deep learning (DL), a branch of Artificial ...

show abstract

Section: Discussionmentioning

confidence: 99%

Convolutional neural network models for cancer type prediction based on gene expression

Chiu²,

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Immuno-active neoantigens' corresponding DNA tend to belong to active chromosomal compartments (compartment A) in some chromosomes; 3) Immuno-active neoantigens' corresponding DNA tend to locate at specific regions in the 3D genome. We believe that the 3D genome information, combining advanced machine learning [27][28][29] and feature selection technologies [30][31][32], will help more precise neoantigen prioritization and discovery, and will eventually benefit precision medicine in cancer immunotherapy.…”

Section: Discussionmentioning

confidence: 99%

A Novel Neoantigen Discovery Approach based on Chromatin High Order Conformation: Mapping the Neoantigen to 3D Genome

Shi

Zhang

Xie

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

High-throughput sequencing technology has yielded reliable and ultra-fast sequencing for DNA and RNA. For tumor cells of cancer patients, when combining the results of DNA and RNA sequencing, one can identify potential neoantigens that stimulate the immune response of the T cell. However, when the somatic mutations are abundant, it is computationally challenging to efficiently prioritize the identified neoantigen candidates according to their ability of activating the T cell immuno-response. Numerous prioritization or prediction approaches have been proposed to address this issue but none of them considers the original DNA loci of the neoantigens from the perspective of 3D genome. Here we retrospect the DNA origins of the immune-positive and non-negative neoantigens in the context of 3D genome and discovered that 1) DNA loci of the immuno-positive neoantigens tend to cluster genome-wise; 2) DNA loci of the immuno-positive neoantigens tend to belong to active chromosomal compartment (compartment A) in some chromosomes; 3). DNA loci of the immuno-positive neoantigens tend to locate at specific regions in the 3D genome. We believe that the 3D genome information will help to increase the precision of neoantigen prioritization and discovery and eventually benefit precision and personalized medicine in cancer immunotherapy.

show abstract

“…Further, different types of somatic mutations data such as point mutation, single-nucleotide variation (SNV), small insertion and deletion (INDEL), copy number aberration (CNA), translocation, and CNVs are used. Literature [51] has observed that somatic mutation data are not only associated with complex diseases but also contribute to the growth of different types of cancers. In particular, literature [5] studied CN changes by comparing healthy and cancer-affected patients, which showed that amplification and deletion of certain genes are more common in certain cancer patients than in healthy people.…”

Section: Related Workmentioning

confidence: 99%

“…Literature [9] used a stacked denoising autoencoder to extract features from the RNAseq data, which are then feed into SVM and shallow ANN to classify malignant or benign tumor of breasts [7]. DeepCNA is another approach proposed for cancer-type prediction based on CNVs and chromatin 3D structure with CNN [51].…”

Section: Related Workmentioning

confidence: 99%

A snapshot neural ensemble method for cancer-type prediction based on copy number variations

Karim

Rahman

Jares

et al. 2019

Neural Comput & Applic

View full text Add to dashboard Cite

An accurate diagnosis and prognosis for cancer are specific to patients with particular cancer types and molecular traits, which needs to address carefully. The discovery of important biomarkers is becoming an important step toward understanding the molecular mechanisms of carcinogenesis in which genomics data and clinical outcomes need to be analyzed before making any clinical decision. Copy number variations (CNVs) are found to be associated with the risk of individual cancers and hence can be used to reveal genetic predispositions before cancer develops. In this paper, we collect the CNVs data about 8000 cancer patients covering 14 different cancer types from The Cancer Genome Atlas. Then, two different sparse representations of CNVs based on 578 oncogenes and 20,308 protein-coding genes, including genomic deletions and duplication across the samples, are prepared. Then, we train Conv-LSTM and convolutional autoencoder (CAE) networks using both representations and create snapshot models. While the Conv-LSTM can capture locally and globally important features, CAE can utilize unsupervised pretraining to initialize the weights in the subsequent convolutional layers against the sparsity. Model averaging ensemble (MAE) is then applied to combine the snapshot models in order to make a single prediction. Finally, we identify most significant CNVs biomarkers using guided-gradient class activation map plus (GradCAM??) and rank top genes for different cancer types. Results covering several experiments show fairly high prediction accuracies for the majority of cancer types. In particular, using protein-coding genes, Conv-LSTM and CAE networks can predict cancer types correctly at least 72.96% and 76.77% of the cases, respectively. Contrarily, using oncogenes gives moderately higher accuracies of 74.25% and 78.32%, whereas the snapshot model based on MAE shows overall 2.5% of accuracy improvement.

show abstract

Cancer type prediction based on copy number aberration and chromatin 3D structure with convolutional neural networks

Cited by 28 publications

References 42 publications

Convolutional neural network models for cancer type prediction based on gene expression

Convolutional neural network models for cancer type prediction based on gene expression

A Novel Neoantigen Discovery Approach based on Chromatin High Order Conformation: Mapping the Neoantigen to 3D Genome

A snapshot neural ensemble method for cancer-type prediction based on copy number variations

Contact Info

Product

Resources

About