基于ResNet的序列模式的多类识别,用于基因组复制的时间分析
Zhen-Ning Yin1, Yu-Hao Zeng1, Feng Gao2,3,4
1Department of Physics, School of Science, Tianjin University, Tianjin, 300072, China.
Interdisciplinary sciences, computational life sciences
|December 12, 2025
概括
本研究介绍了RT-Predictor,这是一种深度学习模型,只使用DNA序列模式准确预测DNA复制时间 (RT) 域. 该框架增强了对基因组稳定性及其与癌症等疾病的联系的理解.
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:
- 精确调节DNA复制时间 (RT) 对基因组稳定性至关重要.
- 识别影响RT的序列模式是具有挑战性的,具有有限的计算工具来准确预测.
- 了解RT动态对于破译基因组不稳定性和包括癌症在内的疾病机制至关重要.
研究的目的:
- 开发基于深度学习的框架RT-Predictor,用于使用序列模式准确预测DNA复制时间域.
- 将人类基因组序列分为四个不同的复制时间域:早期复制域 (ERD),下移转区 (DTZ),晚期复制域 (LRD) 和上移转区 (UTZ).
- 研究特定序列模式,复制起源 (ORIs) 和与DNA损伤修复机制的潜在联系之间的关系.
主要方法:
- 使用深度学习框架,特别是残余网络 (ResNet),用于序列模式分类.
- 从DNA序列中提取了384个特征,包括基于位置和频率的特征,以定义复制动态.
- 进行全基因组RT预测,分析复制时间域的分布和特征.
主要成果:
- 通过仅使用DNA序列模式,RT-Predictor实现了74.58%的准确性,0.6612MCC,0.7458F1得分和0.7457回忆.
- 该模型成功地解决了人类基因组中复杂的DNA复制时间模式.
- 全基因组分析表明,复制起源主要在早期的S阶段启动复制,这表明基于序列模式的调节.
结论:
- 深度学习有效地解码了DNA复制时间中的序列模式的监管意义.
- RT-Predictor为分析复制动态和理解基因组稳定性提供了一个强大的工具.
- 这些发现为基因组稳定的分子基础及其在疾病,特别是癌症中的破坏提供了关键的见解.
相关概念视频
Chromosome Replication
10.4K
Before a cell can divide, it must accurately replicate all of its chromosomes, including the DNA and its associated histone and non-histone proteins. This process begins at numerous origins of replication during the S phase of the cell cycle in each of a cell’s chromosomes simultaneously. Certain nucleotides can act as origins of replication, but these sequences are not well defined - especially in complex, multi-cellular, eukaryotic species. The length of DNA that spans an origin...
10.4K
DNA Replication
58.2K
DNA replication involves the separation of the two strands of the double helix, with each strand serving as a template from which the new complementary strand is copied. After replication, each double-stranded DNA includes one parental or “old” strand and one “new” strand. This is known as semiconservative replication. The resulting DNA molecules have the same sequence and are divided equally into the two daughter cells.
Replication in Prokaryotes
DNA replication...
Replication in Prokaryotes
DNA replication...
58.2K
The Replisome
37.9K
DNA replication is carried out by a large complex of proteins that act in a coordinated matter to achieve high-fidelity DNA replication. Together this complex is known as the DNA replication machinery or the replisome.
The synthesis of the leading and lagging strands is a highly coordinated process. To explain this, the “Trombone model” was proposed by Bruce Alberts in 1980. The DNA loop formation starts when a primer is synthesized on the parent lagging strand. The loop grows with...
The synthesis of the leading and lagging strands is a highly coordinated process. To explain this, the “Trombone model” was proposed by Bruce Alberts in 1980. The DNA loop formation starts when a primer is synthesized on the parent lagging strand. The loop grows with...
37.9K
The Replisome
9.6K
9.6K
Next-generation Sequencing
97.6K
The first human genome sequencing project cost $2.7 billion and was declared complete in 2003, after 15 years of international cooperation and collaboration between several research teams and funding agencies. Today, with the advent of next-generation sequencing technologies, the cost and time of sequencing a human genome have dropped over 100 fold.
Next-Generation Sequencing Methods
Although all next-generation methods use different technologies, they all share a set of standard features....
Next-Generation Sequencing Methods
Although all next-generation methods use different technologies, they all share a set of standard features....
97.6K
Chromosome Structure
25.8K
A functional eukaryotic chromosome must contain three elements: a centromere, telomeres, and numerous origins of replication.
The centromere is a DNA sequence that links sister chromatids. This is also where kinetochores, protein complexes to which spindle microtubules attach, are constructed after the chromosome is replicated. The kinetochores allow the spindle microtubules to move the chromosomes within the cell during cell division.
Telomeres consist of non-coding repetitive nucleotide...
The centromere is a DNA sequence that links sister chromatids. This is also where kinetochores, protein complexes to which spindle microtubules attach, are constructed after the chromosome is replicated. The kinetochores allow the spindle microtubules to move the chromosomes within the cell during cell division.
Telomeres consist of non-coding repetitive nucleotide...
25.8K


