当前位置: X-MOL 学术bioRxiv. Bioinform. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
GuidingNet: revealing transcriptional cofactor and predicting binding for DNA methyltransferase by network regularization
bioRxiv - Bioinformatics Pub Date : 2020-06-02 , DOI: 10.1101/2020.06.02.129445
Lixin Ren , Caixia Gao , Zhana Duren , Yong Wang

The DNA methyltransferases (DNMTs) (DNMT3A, DNMT3B, and DNMT3L) are primarily responsible for the establishment of genomic locus-specific DNA methylation patterns, which play an important role in gene regulation and animal development. However, this important protein family’s binding mechanism, i.e., how and where the DNMTs bind to genome, is still missing in most tissues and cell lines. This motivates us to explore DNMTs and TF’s cooperation and develop a network regularized logistic regression model, GuidingNet, to predict DNMTs’ genome-wide binding by integrating gene expression, chromatin accessibility, sequence, and protein-protein interaction data. GuidingNet accurately predicted methylation experimental data validated DNMTs’ binding, outperformed single data source based method and sparsity regularized methods, and performed well in within and across tissue prediction for several DNMTs in both human and mouse. Importantly, GuidingNet can reveal transcription co-factors assisting DNMTs for methylation establishment. This provides biological understanding in the DNMTs' binding specificity in different tissues and demonstrate the advantage of network regularization. In addition, GuidingNet achieves good performance for chromatin regulators’ binding other than DNMTs and serves as a useful method for studying chromatin regulator binding and function. The GuidingNet is freely available at https://github.com/AMSSwanglab/GuidingNet.

中文翻译:

GuidingNet:通过网络规则化揭示转录辅因子并预测DNA甲基转移酶的结合

DNA甲基转移酶(DNMT)(DNMT3A,DNMT3B和DNMT3L)主要负责建立基因组基因座特异性DNA甲基化模式,这在基因调控和动物发育中起重要作用。然而,在大多数组织和细胞系中仍然缺少这种重要的蛋白质家族的结合机制,即DNMT如何以及在何处与基因组结合。这激励我们探索DNMT和TF的合作,并开发网络正规化逻辑回归模型GuidingNet,以通过整合基因表达,染色质可及性,序列和蛋白质-蛋白质相互作用数据来预测DNMT的全基因组结合。GuidingNet准确预测了甲基化实验数据,从而验证了DNMT的结合,优于基于单一数据源的方法和稀疏正则化方法,在人和小鼠的几种DNMT的组织内和跨组织预测中表现良好。重要的是,GuidingNet可以揭示辅助DNMT进行甲基化建立的转录辅因子。这提供了不同组织中DNMT结合特异性的生物学理解,并证明了网络规则化的优势。此外,除DNMT外,GuidingNet还具有很好的染色质调节剂结合性能,是研究染色质调节剂结合和功能的有用方法。GuidingNet可从https://github.com/AMSSwanglab/GuidingNet免费获得。这提供了不同组织中DNMT结合特异性的生物学理解,并证明了网络规则化的优势。此外,除DNMT以外,GuidingNet还对染色质调节剂的结合具有良好的性能,并且是研究染色质调节剂的结合和功能的有用方法。GuidingNet可从https://github.com/AMSSwanglab/GuidingNet免费获得。这提供了不同组织中DNMT结合特异性的生物学理解,并证明了网络规则化的优势。此外,除DNMT以外,GuidingNet还对染色质调节剂的结合具有良好的性能,并且是研究染色质调节剂的结合和功能的有用方法。GuidingNet可从https://github.com/AMSSwanglab/GuidingNet免费获得。
更新日期:2020-06-02
down
wechat
bug