X-LDA:一种可解释和基于知识的异质图形学习框架,用于LncRNA-疾病关联预测
Yangkun Cao1, Jun Xiao2, Nan Sheng2
1School of Artificial Intelligence, Jilin University, Changchun, 130012, China.
Computers in biology and medicine
|November 4, 2024
概括
我们开发了X-LDA,这是一种可解释的计算框架,用于预测长非编码RNA与疾病的关联. 这种方法增强了对基因调节和疾病机制的理解,优于现有的方法.
科学领域:
- 计算生物学是一种计算生物学.
- 基因组学就是基因组学.
- 生物信息学是一种生物信息学.
背景情况:
- 长非编码RNAs (lncRNAs) 在基因表达调节和表观遗传修饰中起着至关重要的作用.
- 识别 lncRNA-疾病关联 (LDA) 对于理解疾病机制至关重要.
- 现有的LDA计算方法往往缺乏解释性,阻碍了生物和医学研究人员的采用.
研究的目的:
- 提出X-LDA,一个可解释和基于知识的异质图形学习框架,用于预测LDA.
- 为LDA预测提供直观的解释,增强信任和理解.
- 提高当前LDA预测方法的性能和可解释性.
主要方法:
- 构建了一个基于知识的异质图,整合了LDAs,lncRNA相似性和疾病相似性.
- 定义了九种图形补丁类型,以捕捉可解释预测的拓关系.
- 使用图形补丁卷积与参数共享和多卷积内核用于特征提取和上下文嵌入.
- 利用集成梯度来对LDA预测进行后期解释.
主要成果:
- 与九种最先进的方法相比,X-LDA显示出更高的性能.
- 实现了0.9891的接收器操作曲线 (AUC) 下的平均面积和0.7907.7的精度回忆曲线 (AUPRC) 下的平均面积.
- 废除研究和可解释性实验证实了X-LDA的稳定性,可学习性,可预测性和可解释性.
- 关于前列腺癌,结直肠癌和乳腺癌的案例研究展示了X-LDA的实际适用性.
结论:
- X-LDA提供了一种可靠和可解释的方法来预测lncRNA与疾病的关联.
- 该框架增强了对基因调节和疾病的生物学洞察力.
- X-LDA的性能和可解释性使其成为基因组学和医学研究人员的宝贵工具.
相关概念视频
lncRNA - Long Non-coding RNAs
8.5K
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...
8.5K
Genome-wide Association Studies-GWAS
12.5K
Genome-wide association studies or GWAS are used to identify whether common SNPs are associated with certain diseases. Suppose specific SNPs are more frequently observed in individuals with a particular disease than those without the disease. In that case, those SNPs are said to be associated with the disease. Chi-square analysis is performed to check the probability of the allele likely to be associated with the disease.
GWAS does not require the identification of the target gene involved in...
GWAS does not require the identification of the target gene involved in...
12.5K


