基于序列预训练的图形神经网络用于预测lncRNA-miRNA关联.
Zixiao Wang1, Shiyang Liang2, Siwei Liu1
1Mohamed bin Zayed University of Artificial Intelligence, Masdar City, UAE.
Briefings in bioinformatics
|August 31, 2023
概括
我们开发了一种新的深度学习模型,SPGNN,用于预测长非编码RNA (lncRNA) -microRNA (miRNA) 相互作用. 这种方法准确地识别了潜在的ceRNA关系,对于理解基因调节和疾病至关重要.
科学领域:
- 基因组学就是基因组学.
- 生物信息学是一种生物信息学.
- 计算生物学 计算生物学
背景情况:
- 微RNAs (miRNAs) 通过准信使RNAs (mRNAs) 来调节基因表达.
- 长非编码RNAs (lncRNAs) 可以作为竞争性内源RNAs (ceRNAs) 起作用,调节miRNA活动并影响基因表达.
- 鉴定lncRNA-miRNA相互作用对于理解基因调节至关重要,但在实验上具有挑战性.
研究的目的:
- 提出一种新的深度学习框架,基于序列预训练的图形神经网络 (SPGNN),用于预测lncRNA-miRNA关联.
- 利用RNA序列和现有的相互作用网络进行准确的预测.
主要方法:
- 利用序列向向量方法进行RNA序列预训练以产生嵌入.
- 在微调阶段使用图形神经网络 (GNN) 来从 lncRNA-miRNA 相互作用的异质图中学习.
- 结合k-mer技术和Doc2vec进行预训练,并使用简单图形卷积网络进行微调.
主要成果:
- 与动物数据集上最先进的方法相比,SPGNN在预测lncRNA-miRNA关联方面表现优异.
- 通过废弃性研究和超参数分析验证了单个成分和参数的有效性.
结论:
- 拟议的SPGNN模型通过整合序列信息和网络拓学,有效地预测lncRNA-miRNA关联.
- 这种方法为探索ceRNA机制及其医疗影响提供了有价值的计算工具.
相关概念视频
02:39lncRNA - Long Non-coding RNAs
8.6K
In humans, more than 80% of the genome gets transcribed. However, only around 2% of the genome codes for proteins. The remaining part produces non-coding RNAs which includes ribosomal RNAs, transfer RNAs, telomerase RNAs, and regulatory RNAs, among other types. A large number of regulatory non-coding RNAs have been classified into two groups depending upon their length – small non-coding RNAs, such as microRNA, which are less than 200 nucleotides in length, and long non-coding RNA...
8.6K
01:22MicroRNAs
3.0K
MicroRNA (miRNA) are short, regulatory RNA transcribed from introns (non-coding regions of a gene) or intergenic regions (stretches of DNA present between genes). Several processing steps are required to form biologically active, mature miRNA. The initial transcript, called primary miRNA (pri-mRNA), base-pairs with itself, forming a stem-loop structure. Within the nucleus, an endonuclease enzyme, called Drosha, shortens the stem-loop structure into hairpin-shaped pre-miRNA. After the pre-miRNA...
3.0K