» Articles » PMID: 31569801

A Shallow Convolutional Learning Network for Classification of Cancers Based on Copy Number Variations

Overview
Journal Sensors (Basel)
Publisher MDPI
Specialty Biotechnology
Date 2019 Oct 2
PMID 31569801
Citations 4
Authors
Affiliations
Soon will be listed here.
Abstract

Genomic copy number variations (CNVs) are among the most important structural variations. They are linked to several diseases and cancer types. Cancer is a leading cause of death worldwide. Several studies were conducted to investigate the causes of cancer and its association with genomic changes to enhance its management and improve the treatment opportunities. Classification of cancer types based on the CNVs falls in this category of research. We reviewed the recent, most successful methods that used machine learning algorithms to solve this problem and obtained a dataset that was tested by some of these methods for evaluation and comparison purposes. We propose three deep learning techniques to classify cancer types based on CNVs: a six-layer convolutional net (CNN6), residual six-layer convolutional net (ResCNN6), and transfer learning of pretrained VGG16 net. The results of the experiments performed on the data of six cancer types demonstrated a high accuracy of 86% for ResCNN6 followed by 85% for CNN6 and 77% for VGG16. The results revealed a lower prediction accuracy for one of the classes (uterine corpus endometrial carcinoma (UCEC)). Repeating the experiments after excluding this class reveals improvements in the accuracies: 91% for CNN6 and 92% for Res CNN6. We observed that UCEC and ovarian serous carcinoma (OV) share a considerable subset of their features, which causes a struggle for learning in the classifiers. We repeated the experiment again by balancing the six classes through oversampling of the training dataset and the result was an enhancement in both overall and UCEC classification accuracies.

Citing Articles

Chromothripsis detection with multiple myeloma patients based on deep graph learning.

Yu J, Chen N, Zheng Z, Gao M, Liang N, Wong K Bioinformatics. 2023; 39(7).

PMID: 37399092 PMC: 10343948. DOI: 10.1093/bioinformatics/btad422.


Transfer learning for non-image data in clinical research: A scoping review.

Ebbehoj A, Thunbo M, Andersen O, Glindtvad M, Hulman A PLOS Digit Health. 2023; 1(2):e0000014.

PMID: 36812540 PMC: 9931256. DOI: 10.1371/journal.pdig.0000014.


Multiclass Cancer Prediction Based on Copy Number Variation Using Deep Learning.

Attique H, Shah S, Jabeen S, Khan F, Khan A, ElAffendi M Comput Intell Neurosci. 2022; 2022:4742986.

PMID: 35720914 PMC: 9203194. DOI: 10.1155/2022/4742986.


Genomic pan-cancer classification using image-based deep learning.

Ye T, Li S, Zhang Y Comput Struct Biotechnol J. 2021; 19:835-846.

PMID: 33598099 PMC: 7848437. DOI: 10.1016/j.csbj.2021.01.010.

References
1.
Mahas A, Potluri K, Kent M, Naik S, Markey M . Copy number variation in archival melanoma biopsies versus benign melanocytic lesions. Cancer Biomark. 2016; 16(4):575-97. DOI: 10.3233/CBM-160600. View

2.
Yuan Y, Shi Y, Su X, Zou X, Luo Q, Feng D . Cancer type prediction based on copy number aberration and chromatin 3D structure with convolutional neural networks. BMC Genomics. 2018; 19(Suppl 6):565. PMC: 6101087. DOI: 10.1186/s12864-018-4919-z. View

3.
Mermel C, Schumacher S, Hill B, Meyerson M, Beroukhim R, Getz G . GISTIC2.0 facilitates sensitive and confident localization of the targets of focal somatic copy-number alteration in human cancers. Genome Biol. 2011; 12(4):R41. PMC: 3218867. DOI: 10.1186/gb-2011-12-4-r41. View

4.
Montgomery S, Goode D, Kvikstad E, Albers C, Zhang Z, Mu X . The origin, evolution, and functional impact of short insertion-deletion variants identified in 179 human genomes. Genome Res. 2013; 23(5):749-61. PMC: 3638132. DOI: 10.1101/gr.148718.112. View

5.
Du W, Elemento O . Cancer systems biology: embracing complexity to develop better anticancer therapeutic strategies. Oncogene. 2014; 34(25):3215-25. DOI: 10.1038/onc.2014.291. View