DL-CNV: A deep learning method for identifying copy number variations based on next generation target sequencing

Yunxiang Zhang; Lvcheng Jin; Bo Wang; Dehong Hu; Leqiang Wang; Pan Li; Junling Zhang; Kai Han; Geng Tian; Dawei Yuan; Jialiang Yang; Wei Tan; Xiaoming Xing; Jidong Lang; Yunxiang Zhang; Lvcheng Jin; Bo Wang; Dehong Hu; Leqiang Wang; Pan Li; Junling Zhang; Kai Han; Geng Tian; Dawei Yuan; Jialiang Yang; Wei Tan; Xiaoming Xing; Jidong Lang

doi:10.3934/mbe.2020011

Mathematical Biosciences and Engineering

2020, Volume 17, Issue 1: 202-215. doi: 10.3934/mbe.2020011

Previous Article Next Article

Research article Special Issues

DL-CNV: A deep learning method for identifying copy number variations based on next generation target sequencing

1.
Weifang People's Hospital, Guang Wen Road, Weifang 261000, China
2.
Weifang Medical University, Bao Tong West Street, Weifang 261053, China
3.
Geneis Beijing Limited Company, Beijing 100102, China
4.
The Affiliated Hospital of Qingdao University, Jiang Su Road, Qingdao 266071, China

Received: 28 April 2019 Accepted: 17 September 2019 Published: 30 September 2019

Copy number variations (CNVs) play an important role in many types of cancer. With the rapid development of next generation sequencing (NGS) techniques, many methods for detecting CNVs of a single sample have emerged: (ⅰ) require genome-wide data of both case and control samples, (ⅱ) depend on sequencing depth and GC content correction algorithm, (ⅲ) rely on statistical models built on CNV positive and negative sample datasets. These make them costly in the data analysis and ineffective in the targeted sequencing data. In this study, we developed a novel alignment-free method called DL-CNV to call CNV from the target sequencing data of a single sample. Specifically, we collected two sets of samples. The first set consists of 1301 samples, in which 272 have CNVs in ERBB2 and the second set is composed of 1148 samples with 63 samples containing CNVs in MET. Finally, we found that a testing AUC of 0.9454 for ERBB2 and 0.9220 for MET. Furthermore, we hope to make the CNV detection could be more accurate with clinical pgold standardq (e.g. FISH) information and provide a new research direction, which can be used as the supplement to the existing NGS methods.
- copy number variation,
- next generation sequencing,
- deep learning,
- convolutional neural network,
- target sequencing
Citation: Yunxiang Zhang, Lvcheng Jin, Bo Wang, Dehong Hu, Leqiang Wang, Pan Li, Junling Zhang, Kai Han, Geng Tian, Dawei Yuan, Jialiang Yang, Wei Tan, Xiaoming Xing, Jidong Lang. DL-CNV: A deep learning method for identifying copy number variations based on next generation target sequencing[J]. Mathematical Biosciences and Engineering, 2020, 17(1): 202-215. doi: 10.3934/mbe.2020011

Related Papers:

Abstract

Copy number variations (CNVs) play an important role in many types of cancer. With the rapid development of next generation sequencing (NGS) techniques, many methods for detecting CNVs of a single sample have emerged: (ⅰ) require genome-wide data of both case and control samples, (ⅱ) depend on sequencing depth and GC content correction algorithm, (ⅲ) rely on statistical models built on CNV positive and negative sample datasets. These make them costly in the data analysis and ineffective in the targeted sequencing data. In this study, we developed a novel alignment-free method called DL-CNV to call CNV from the target sequencing data of a single sample. Specifically, we collected two sets of samples. The first set consists of 1301 samples, in which 272 have CNVs in ERBB2 and the second set is composed of 1148 samples with 63 samples containing CNVs in MET. Finally, we found that a testing AUC of 0.9454 for ERBB2 and 0.9220 for MET. Furthermore, we hope to make the CNV detection could be more accurate with clinical pgold standardq (e.g. FISH) information and provide a new research direction, which can be used as the supplement to the existing NGS methods.

References

[1]	S. A. McCarroll and D. M. Altshuler, Copy-number variation and association studies of human disease, Nat. Genet., 39(2007), S37-42.
[2]	P. Liu, C. M. Carvalho, P. J. Hastings, et al., Mechanisms for recurrent and complex human genomic rearrangements, Curr. Opin. Genet. Dev., 22(2012), 211-220.
[3]	A. P. de Koning, W. Gu, T. A. Castoe, et al., Repetitive elements may comprise over two-thirds of the human genome, Plos Genet.,7 (2011), e1002384. doi: 10.1371/journal.pgen.1002384
[4]	M. Zarrei, J. R. MacDonald, D. Merico, et al., A copy number variation map of the human genome, Nat. Rev. Genet., 16(2015), 172-183. doi: 10.1038/nrg3871
[5]	J. L. Freeman, G. H. Perry, L. Feuk, et al., Copy number variation: new insights in genome diversity, Genome Res., 16(2006), 949-961.
[6]	S. F. Chin, A. E. Teschendorff, J. C. Marioni, et al., High-resolution aCGH and expression profiling identifies a novel genomic subtype of ER negative breast cancer, Genome Biol., 8(2007), R215.
[7]	D. He, N. Furlotte and E. Eskin, Detection and reconstruction of tandemly organized de novo copy number variations, BMC Bioinf., 11(2010), S12.
[8]	G. Klambauer, K. Schwarzbauer, A. Mayr, et al., cn.MOPS: Mixture of Poissons for discovering copy number variations in next-generation sequencing data with a low false discovery rate, Nucleic Acids Res., 40 (2012), e69. doi: 10.1093/nar/gks003
[9]	E. Talevich, A. H. Shain, T. Botton, et al., CNVkit: Genome-Wide Copy Number Detection and Visualization from Targeted DNA Sequencing, PLoS Comput. Biol., 12(2016), e1004873.
[10]	A. Abyzov, A. E. Urban, M. Snyder, et al., CNVnator: An approach to discover, genotype, and characterize typical and atypical CNVs from family and population genome sequencing, Genome Res., 21(2011), 974-984.
[11]	V. Boeva, T. Popova, K. Bleakley, et al., Control-FREEC: Atool for assessing copy number and allelic content using next-generation sequencing data, Bioinf., 28(2012), 423-425.
[12]	G. Onsongo, L. B. Baughn, M. Bower, et al., CNV-RF Is a Random Forest-Based Copy Number Variation Detection Method Using Next-Generation Sequencing, J. Mol. Diagn., 18(2016), 872-881.
[13]	C. Wang, J. M. Evans, A. V. Bhagwate, et al., PatternCNV: Aversatile tool for detecting copy number changes from exome sequencing data, Bioinf., 30(2014), 2678-2680.
[14]	J. Budczies, N. Pfarr, A. Stenzinger, et al., Ioncopy: Anovel method for calling copy number alterations in amplicon sequencing data including significance assessment, Oncotarget, 7 (2016), 13236-13247.
[15]	Y. L. Cun, L. Bottou, Y. Bengio, et al., Gradient-Based Learning Applied to Document Recognition, Proc. IEEE., 86 (1998), 2278-2324.
[16]	J. Zhou and O. G. Troyanskaya, Predicting effects of noncoding variants with deep learning-based sequence model, Nat. Methods,12 (2015), 931-934.
[17]	R. Poplin, D. Newburger, J. Dijamco, et al., Creating a universal SNP and small indel variant caller with deep neural networks, BioRxiv, (2016), 092890.
[18]	M. X. Sliwkowski, J. A. Lofgren, G. D. Lewis, et al., Nonclinical studies addressing the mechanism of action of trastuzumab (Herceptin), Semin Oncol., 26(1999), 60-70.
[19]	S. Ahn, M. Hong, M. Van Vrancken, et al., A nCounter CNV Assay to Detect HER2 Amplification: A Correlation Study with Immunohistochemistry and In Situ Hybridization in Advanced Gastric Cancer, Mol. Diagn. Ther., 20(2016), 375-383.
[20]	F. Sircoulomb, I. Bekhouche, P. Finetti, et al., Genome profiling of ERBB2-amplified breast cancers, BMC Cancer,10 (2010), 539.
[21]	S. Kim, T. M. Kim, D. W. Kim, et al., Acquired Resistance of MET-Amplified Non-small Cell Lung Cancer Cells to the MET Inhibitor Capmatinib, Cancer Res. Treat., 5 (2019), 951-962.
[22]	N. Pfarr, R. Penzel, F. Klauschen, et al., Copy number changes of clinically actionable genes in melanoma, non-small cell lung cancer and colorectal cancer-A survey across 822 routine diagnostic cases, Genes Chromosomes Cancer, 55 (2016), 821-833.
[23]	F. Zare, M. Dow, N. Monteleone, et al., An evaluation of copy number variation detection tools for cancer using whole exome sequencing data, BMC Bioinf., 18(2017), 286.
[24]	H. Li and R. Durbin, Fast and accurate short read alignment with Burrows-Wheeler transform, Bioinf., 25(2009), 1754-1760.
[25]	H. Li, B. Handsaker, A. Wysoker, et al., The Sequence Alignment/Map format and SAMtools, Bioinf., 25(2009), 2078-2079.
[26]	M. Abadi, P. Barham, J. Chen, et al., TensorFlow: A system for large-scale machine learning, 12th Symposium on Operating Systems Design and Implementation (OSDI), 2016, 265-283. Available from: https://www.usenix.org/conference/osdi16/technical-sessions/presentation/abadi.

Reader Comments

Your name:*

Email:*
© 2020 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)