变分自编码器对甲基化缺失数据的填补

doi:10.3778/j.issn.1002-8331.2109-0135

摘要/Abstract

摘要： 针对高通量测序技术因各种原因导致的DNA甲基化测序数据中包含部分缺失值的问题。提出一种基于变分自编码器的DNA甲基化缺失数据填补模型VAE-MethImp。VAE-MethImp是一种深度隐含空间生成模型，由编码层、隐含层和解码层组成，拥有强大的重构输入数据能力。编码层进行均值和方差的推断;隐含层是通过编码层输出的均值和方差计算出的输入数据的专属正态分布;解码层对隐含层包含的特征进行解码生成重构后的数据。通过在肺癌和乳腺癌上的填补实验证明，VAE-MethImp提取的特征更具信息性。在填补精度上，VAE-MethImp比对照方法（均值（Mean）、最近邻（KNN）、主成分分析（PCA）和奇异值分解（SVD））中最优的SVD提升了4.8%。生存分析实验结果显示VAE-MethImp填补的数据具有更好的预测性，同时也证明DNA甲基化与癌症的生存存在直接关联。

关键词: 深度学习, 变分自编码器, DNA甲基化, 数据填补, 生存分析

Abstract: High-throughput sequencing technology is an important method for studying DNA methylation. Due to experimental and technical limitations, DNA methylation sequencing data contains some missing values. To solve the problem of missing values, the VAE-MethImp model based on variational auto-encoder for DNA methylation missing data imputation is proposed. VAE-MethImp is composed of an encoder layer, a hidden layer and a decoder layer. It is a deeply hidden space generation model with a powerful ability to reconstruct input data. The encoder layer infers the mean and variance; the hidden layer is the exclusive normal distribution of the input data calculated from the mean and variance output by the encoder layer; the decoder layer decodes the information contained in the hidden variables to generate reconstructed data. The imputation experiments on lung cancer and breast cancer prove that the features extracted by the VAE are more informative. The imputation accuracy of the VAE model is 4.8% higher than the optimal SVD among the four traditional methods, K-nearest neighbor（KNN）, principal component analysis（PCA）, and singular value decomposition（SVD）. The survival analysis experiment results show that the data imputed by the VAE has better predictability, and it also proves that DNA Methylation is directly related to cancer survival.

Key words: deep learning, variational auto-encoder, DNA methylation, data imputation, survival analysis

王新峰, 黄伟. 变分自编码器对甲基化缺失数据的填补[J]. 计算机工程与应用, 2022, 58(12): 149-154.

WANG Xinfeng, HUANG Wei. Data Imputation for Methylation by Variational Auto-Encoder[J]. Computer Engineering and Applications, 2022, 58(12): 149-154.

参考文献

[1] FRANCIS R C.Epigenetics：the ultimate mystery of inheritance[J].Journal of the Association of Genetic Technologists，2011，1：15-24.
[2] YE P，LUAN Y，CHEN K，et al.MethSMRT：an integrative database for DNA N6-methyladenine and N4-methylcytosine generated by single-molecular real-time sequencing[J].Nucleic Acids Research，2016，45（1）：85-89.
[3] KULIS M，ESTELLER M.DNA methylation and cancer[J].Advances in Genetics，2010，70（22）：27-56.
[4] GERD P.Defining driver DNA methylation changes in human cancer[J].International Journal of Molecular Sciences，2018，19（4）：1166-1180.
[5] ZHANG G，HUANG K C，XU Z，et al.Across-platform imputation of DNA methylation levels incorporating nonlocal information using penalized functional regression[J].Genetic Epidemiology，2016，40（4）：333-340.
[6] TROYANSKAYA O，CANTOR M，SHERLOCK G，et al.Missing value estimation methods for DNA microarrays[J].Bioinformatics，2001，17（6）：520-525.
[7] 郝胜轩，宋宏，周晓锋.基于近邻噪声处理的KNN缺失数据填补算法[J].计算机仿真，2014，31（7）：264-268.
HAO S X，SONG H，ZHOU X F.Predicting missing values with KNN based on the elimination of neighbor noise[J].Computer Simulation，2014，31（7）：264-268.
[8] YING S，DENG G.Kernel PCA regression for missing data estimation in DNA microarray analysis[C]//IEEE International Symposium on Circuits & Systems，2009.
[9] HUSSON F，JOSSE J.Handling missing values in exploratory multivariate data analysis methods[J].Journal de La société Fran?aise de Statistique & Revue de Statistique Appliquée，2012，153（2）：1-21.
[10] LECUN Y，BENGIO Y，HINTON G.Deep learning[J].Nature，2015，521（7553）：436-444.
[11] 史加荣，马媛媛.深度学习的研究进展与发展[J].计算机工程与应用，2018，54（10）：1-10.
SHI J R，MA Y Y.Research progress and development of deep learning[J].Computer Engineering and Applications，2018，54（10）：1-10.
[12] 王山海，景新幸，杨海燕.基于深度学习神经网络的孤立词语音识别的研究[J].计算机应用研究，2015，32（8）：2289-2291.
WANG S H，JING X X，YANG H Y.Study of isolated speech recognition based on deep learning neural networks[J].Application Research of Computers，2015，32（8）：2289-2291.
[13] 张顺，龚怡宏，王进军.深度卷积神经网络的发展及其在计算机视觉领域的应用[J].计算机学报，2019，42（3）：453-482.
ZHANG S，GONG Y H，WANG J J.The development of deep convolution neural network and its applications on computer vision[J].Chinese Journal of Computers，2019，42（3）：453-482.
[14] 吴玉佳，李晶，宋成芳，等.基于高效用神经网络的文本分类方法[J].电子学报，2020，48（2）：279-284.
WU Y J，LI J，SONG C F，et al.High utility neural networks for text classification[J].Acta Electronica Sinica，2020，48（2）：279-284.
[15] HEFFERNAN R，PALIWAL K，LYONS J，et al.Improving prediction of secondary structure，local backbone angles，and solvent accessible surface area of proteins by iterative deep learning[J].Scientific Reports，2015，5：11476.
[16] CHEN J，ZHENG S，ZHAO H，et al.Structure-aware protein solubility prediction from sequence through graph convolutional network and predicted contact map[J].Journal of Chemin for Matics，2021，13（1）.
[17] SENIOR A W，EVANS R，JUMPER J，et al.Improved protein structure prediction using potentials from deep learning[J].Nature，2020，577（7792）：706-710.
[18] CHING T，HIMMELSTEIN D S.Opportunities and obstacles for deep learning in biology and medicine[J].Journal of the Royal Society Interface，2018，15（141）：20170387.
[19] ZHENG S，LI Y，CHEN S，et al.Predicting drug-protein interaction using quasi-visual question answering system[J].Nature Machine Intelligence，2020，2（2）：134-140.
[20] ZHENG S，RAO J，ZHANG Z，et al.Predicting ret-rosynthetic reactions using self-corrected transformer neural networks[J].Journal of Chemical Information and Modeling，2019，60（1）：47-55.
[21] WAY G P，GREENE C S.Extracting a biologically relevant latent space from cancer transcriptomes with variational autoencoders[C]//Pacific Symposium on Biocomputing，2018，23：80-91.
[22] TITUS A J，WILKINS O M，BOBAK C A，et al.Un-supervised deep learning with variational autoencoders applied to breast tumor genome-wide DNA methylation data with biologic feature extraction[J].bioRxiv，2018，4：3376-3383.
[23] LV X，CHEN Z，LU Y，et al.An end-to-end ox-ford nanopore basecaller using convolution-augmented transformer[C]//2020 IEEE International Conference on Bioinformatics and Biomedicine（BIBM），2020：17.
[24] TIAN T，WAN J，SONG Q，et al.Clustering single-cell RNA-seq data with a model-based deep learning approach[J].Nature Machine Intelligence，2019，1（4）：191-198.
[25] LOPEZ R，REGIER J，COLE M B，et al.Deep generative modeling for single-cell transcriptomics[J].Nature Methods，2018，15（12）：1053-1058.
[26] ZENG Y，ZHOU X，RAO J，et al.Accurately clustering single-cell RNA-seq data by capturing structural relations between cells through graph convolutional network[C]//2020 IEEE International Conference on Bioinformatics and Biomedicine（BIBM），2020：16.
[27] ZHOU X，CHAI H，ZENG Y，et al.scAdapt：virtual adversarial domain adaptation network for single cell RNA-seq data classification across platforms and species[J].Briefings in Bioinformatics，2021，1（18）：427083.
[28] ZHOU X，CHAI H，ZHAO H，et al.Imputing missing RNA-sequencing data from DNA methylation by using a transfer learning-based neural network[J].GigaScience，2020，9（7）.
[29] KINGMA D P，WELLING M.Auto-encoding variational bayes[J].arXiv：1312.6114，2013.
[30] 翟正利，梁振明，周炜，等.变分自编码器模型综述[J].计算机工程与应用，2019，55（3）：1-9.
ZHAI Z L，LIANG Z M，ZHOU W，et al.Research overview of variational auto-encoders models[J].Computer Engineering and Applications，2019，55（3）：1-9.
[31] WEI L，JIN Z，YANG S，et al.TCGA-assembler 2：software pipeline for retrieval and processing of TCGA/CPTAC data[J].Bioinformatics，2017，34（9）：1615-1617.
[32] VAN BELLE V，PELCKMANS K，VAN HUFFEL S，et al.Support vector methods for survival analysis：a comparison between ranking and regression approaches[J].Artificial Intelligence in Medicine，2011，53（2）：107-118.
[33] FRIEDMAN J，HASTIE T，TIBSHIRANI R.Regularization paths for generalized linear models via coordinate descent[J].Journal of Statistical Software，2010，33（1）：1-22.