HEAP:一个基于任务适应性的可解释的深度学习框架,用于增强活动预测
Yuhang Liu1, Zixuan Wang2, Hao Yuan1
1School of Computer Science, Chengdu University of Information Technology, 610225, Chengdu, China.
Briefings in bioinformatics
|August 4, 2023
概括
新的深度学习工具HEAP使用DNA序列和表观遗传数据准确预测增强器活动和语法. 这种可解释的框架增强了对跨细胞类型的基因调节的理解.
科学领域:
- 基因组学就是基因组学.
- 计算生物学 计算生物学
- 分子生物学分子生物学
背景情况:
- 增强剂是控制细胞类型特定基因表达的关键 cis 调节元素.
- 预测增强器活动和理解它们的语法仍然是基因组学中的重大挑战.
研究的目的:
- 引入HEAP (高分辨率增强器活动预测),一个可解释的深度学习框架.
- 为了准确预测增强剂的活性,并在各种细胞类型中探索增强剂语法.
主要方法:
- HEAP使用基于语法的深度学习方法,包括DNA序列和表观遗传修饰.
- 一种新的两步多任务学习方法,任务自适应参数共享 (TAPS),用于高效的细胞类型特定预测.
- 后期解释方法为预测机制提供了洞察力.
主要成果:
- 与现有方法相比,HEAP在增强器预测方面表现优越.
- 任务适应性参数共享 (TAPS) 方法被证明是有效的,特别是在有限的培训数据的情况下.
- HEAP的可解释性质为模型决策和增强语法提供了洞察力.
结论:
- HEAP是预测增强器活动和理解增强器语法的一个有价值的工具.
- 该框架为管理基因调节的复杂机制提供了更深入的见解.
- HEAP在计算基因组学和调控元件分析领域取得了进展.
相关概念视频
RNA Polymerase II Accessory Proteins
9.2K
Proteins that regulate transcription can do so either via direct contact with RNA Polymerase or through indirect interactions facilitated by adaptors, mediators, histone-modifying proteins, and nucleosome remodelers. Direct interactions to activate transcription is seen in bacteria as well as in some eukaryotic genes. In these cases, upstream activation sequences are adjacent to the promoters, and the activator proteins interact directly with the transcriptional machinery. For example, in...
9.2K
Improving Translational Accuracy
11.6K
Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...
11.6K
Co-activators and Co-repressors
2.4K
2.4K
Eukaryotic Transcription Activators
11.1K
Transcription activators are proteins that promote the transcription of genes from DNA to RNA. In most cases, these proteins contain two separate domains ‒ a domain that binds to DNA and a domain for activating transcription; however, in some cases, a single domain is responsible for both binding and activation of transcription, as seen in the glucocorticoid receptor and MyoD.
The binding domains are capable of recognizing and interacting with regulatory sequences on the DNA. These...
The binding domains are capable of recognizing and interacting with regulatory sequences on the DNA. These...
11.1K
Neural Regulation
39.5K
Digestion begins with a cephalic phase that prepares the digestive system to receive food. When our brain processes visual or olfactory information about food, it triggers impulses in the cranial nerves innervating the salivary glands and stomach to prepare for food.
39.5K


