创新的数据增强策略,用于对具有有限基因表征的生物数据集进行深度学习,重点关注叶绿体基因组
Mohammad Ali Abbasi-Vineh1, Shirin Rouzbahani1, Kaveh Kavousi2
1Department of Agricultural Biotechnology, Tarbiat Modares University (TMU), Tehran, 1497713111, Iran.
Scientific reports
|July 27, 2025
概括
在欧米克学中数据稀缺是深度学习 (DL) 的挑战. 这项研究引入了新的数据增强策略,包括滑窗和k-mer方法,以提高DL模型在有限的生物序列数据上的性能.
科学领域:
- 生物信息学是一种生物信息学.
- 计算生物学 计算生物学
- 基因组学就是基因组学.
背景情况:
- 数据稀缺是将深度学习 (DL) 应用于omics数据集的一个主要限制,特别是来自遗传约束生物或特定生物背景的序列.
- 现有的方法在有限的数据上扎,导致模型训练中的过拟合和非代表性的序列变化等问题.
研究的目的:
- 开发和验证新的数据增强策略,以克服奥米克数据集中的数据稀缺性,用于深度学习应用.
- 为了提高在有限的生物序列数据上训练的深度学习模型的性能和稳定性.
主要方法:
- 采用滑窗技术,生成重叠的次序,具有受控的重叠和共享的核酸特征.
- 一种混合的卷积神经网络 (CNN) 和长期短期记忆 (LSTM) 模型被应用于来自微藻和植物叶绿体的基因和蛋白质的增强数据集.
- 为未标记的数据集开发了一个基于k-mer的数据增强策略,以支持无监督分析.
主要成果:
- 拟议的数据增强策略使DL方法能够有效地应用于具有有限数据的omics数据集.
- 观察到模型性能显著改善,减轻了过拟合和非代表性的序列变化.
- 增强过程在不同的生物数据库中显示出高度适应性.
结论:
- 开发的数据增强策略提供了强大的解决方案,以优化深度学习模型训练,使用有限的OMIC数据.
- 这些方法提高了分析遗传约束生物数据集的潜力,并提高了生物信息学DL模型的可靠性.
- 这些战略提供了一个灵活的框架,用于解决各种生物序列分析任务中的数据稀缺问题.
相关概念视频
Export of Mitochondrial and Chloroplast Genes
3.8K
A eukaryotic cell can have up to three different types of genetic systems: nuclear, mitochondrial, and chloroplast. During evolution, organelles have exported many genes to the nucleus; this transfer is still ongoing in some plant species. Approximately 18% of the Arabidopsis thaliana nuclear genome is thought to be derived from the chloroplast’s cyanobacterial ancestor, and around 75% of the yeast genome derived from the mitochondria’s bacterial ancestor. This export has occurred...
3.8K
Evolutionary Relationships through Genome Comparisons
6.2K
Genome comparison is one of the excellent ways to interpret the evolutionary relationships between organisms. The basic principle of genome comparison is that if two species share a common feature, it is likely encoded by the DNA sequence conserved between both species. The advent of genome sequencing technologies in the late 20th century enabled scientists to understand the concept of conservation of domains between species and helped them to deduce evolutionary relationships across diverse...
6.2K
Comparing Mitochondrial, Chloroplast, and Prokaryotic Genomes
13.5K
The present-day mitochondrial and chloroplast genomes have retained some of the characteristics of their ancestral prokaryotes and also have acquired new attributes during their evolution within eukaryotic cells. Like prokaryotic genomes, mitochondrial and chloroplast genomes neither bind with histone-like proteins nor show complex packaging into chromosome-like structures, as observed in eukaryotes. Unlike mitotic cell divisions observed in eukaryotic cells, mitochondria and chloroplasts...
13.5K
The Anatomy of Chloroplasts
5.7K
Green algae and plants, including green stems and unripe fruit, harbor specialized organelles called chloroplasts to carry out photosynthesis. They coordinate both stages of photosynthesis — the light-dependent reactions and the light-independent reactions. The light-dependent reactions use sunlight to release oxygen and produce chemical energy in the form of ATP and NADPH, and the light-independent reactions capture CO2 and use ATP and NADPH to produce sugar.
Structure of...
Structure of...
5.7K
Anatomy of Chloroplasts
111.5K
Green algae and plants, including green stems and unripe fruit, harbor chloroplasts—the vital organelles where photosynthesis takes place. In plants, the highest density of chloroplasts is found in the mesophyll cells of leaves.
111.5K
Genome Size and the Evolution of New Genes
8.3K
While every living organism has a genome of some kind (be it RNA, or DNA), there is considerable variation in the sizes of these blueprints. One major factor that impacts genome size is whether the organism is prokaryotic or eukaryotic. In prokaryotes, the genome contains little to no non-coding sequence, such that genes are tightly clustered in groups or operons sequentially along the chromosome. Conversely, the genes in eukaryotes are punctuated by long stretches of non-coding sequence.
8.3K


