在从未吸烟者中使用深度学习对肺癌的基因组表征
Monjoy Saha1, Thi-Van-Trinh Tran1, Praphulla M S Bhawsar1
1Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, USA.
概括
这项研究引入了一种新的深度学习模型,用于从组织学图像中预测从未吸烟者肺腺癌 (NS-LUAD) 的遗传特征. 该模型准确地识别了11种分子变化,有助于为这种独特的肺癌子组提供个性化的治疗.
科学领域:
- 计算病理学计算病理学
- 在瘤学瘤学.
- 生物信息学是一种生物信息学.
背景情况:
- 深度学习显示出从整个幻灯片图像 (WSIs) 中推断遗传特征的前景.
- 从未吸烟者的肺腺癌 (NS-LUAD) 是分子上不同的,但目前的深度学习模型还没有充分研究.
- 现有的模型通常集中在具有有限分子范围和可变性能的吸烟者群体上.
研究的目的:
- 开发和验证一个定制的深度卷积神经网络,用于从H&E染色WSIs中对NS-LUAD中的分子变化的多标签分类.
- 为了能够同时预测16个分子变化,包括突变,放大,融合和副本数量的变化.
- 优化模型以减少计算复杂性,同时保持预测准确度.
主要方法:
- 开发了一个基于ResNet50的深度卷积神经网络,并进行了结构修改 (简化余块,选择性快捷方式,西格分类头).
- 该模型在495个来自Sherlock-Lung研究的WSI上进行了训练和评估,使用70%的训练,10%的内部测试和30%的持久验证组分.
- 使用接收器操作特征曲线 (AUROC) 下的面积来预测16个分子特征来评估性能.
主要成果:
- 该模型实现了11个特征的高AUROC值 (0.84-0.93),包括EGFR,KRAS,TP53突变和ALK融合.
- 对于瘤突变负担 (AUROC=0.67) 和KRAS热点突变 (例如,p.G12C:AUROC=0.74) 的表现适度.
- 定制模型在同一数据集上的大多数特征上优于Inception-v3等既定架构.
结论:
- 开发的深度学习模型显示了从WSIs直接预测NS-LUAD分子变化的巨大潜力.
- 这种方法可以促进分子测试的分类,并为从未吸烟的肺癌患者提供精确的治疗策略.
- 进一步优化可能会提高其在NS-LUAD临床决策中的实用性.
相关概念视频
Genomics
40.7K
Genomics is the science of genomes: it is the study of all the genetic material of an organism. In humans, the genome consists of information carried in 23 pairs of chromosomes in the nucleus, as well as mitochondrial DNA. In genomics, both coding and non-coding DNA is sequenced and analyzed. Genomics allows a better understanding of all living things, their evolution, and their diversity. It has a myriad of uses: for example, to build phylogenetic trees, to improve productivity and...
40.7K
Genomic Imprinting and Inheritance
37.2K
Diploid organisms inherit genetic material through chromosomes from both parents. Copies of the same gene are known as alleles. In most cases, both alleles are simultaneously expressed and allow various cellular processes to function optimally. If one of the alleles is missing or mutated, the expression of the other allele can compensate; however, this is not true for all genes.
The expression of some genes depends on which parent passed the gene to the offspring, through a phenomenon known as...
The expression of some genes depends on which parent passed the gene to the offspring, through a phenomenon known as...
37.2K
Lung Capacity
56.4K
The air in the lungs is measured in volumes and capacities. Lung volume measures reflect the amount of air taken in, released, or left over after a lung function, like a single inhalation. Lung capacity measures are sums of two or more lung volume measures.
56.4K
Avoidance Learning and Learned Helplessness
2.6K
Avoidance learning and learned helplessness are critical concepts in understanding behavioral responses to negative stimuli.
Avoidance learning occurs when an organism learns that a specific behavior can prevent an unpleasant outcome. For example, a student who receives a bad grade may start studying harder to avoid future poor grades. This behavior persists even when the negative outcome is no longer present. Avoidance learning is powerful because it maintains behavior in the absence of the...
Avoidance learning occurs when an organism learns that a specific behavior can prevent an unpleasant outcome. For example, a student who receives a bad grade may start studying harder to avoid future poor grades. This behavior persists even when the negative outcome is no longer present. Avoidance learning is powerful because it maintains behavior in the absence of the...
2.6K
Genome Size and the Evolution of New Genes
9.1K
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.
9.1K
Comparing Mitochondrial, Chloroplast, and Prokaryotic Genomes
16.3K
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...
16.3K


