基于基因芯片技术的转录组表达研究

发布时间：2018-07-09 14:54

本文选题：基因表达 + 选择性剪切　；参考：《南京航空航天大学》2015年硕士论文

【摘要】：研究表明高等真核生物中普遍存在选择性剪切现象,选择性剪接的异常会引起转录组异常而导致某些疾病的发生,如帕金森病、早老性痴呆、强直性肌营养不良等。因此,转录组表达的研究是近年来生物医学领域的一个研究热点,基于基因芯片的基因与异构体表达水平的分析研究为揭示转录组表达的变化情况提供了一种可行的研究方式。Affmetrix公司制备的外显子芯片和HTA2.0芯片是转录组研究中常用的基因芯片实验技术。在转录组研究中,转录组表达水平计算的精度影响后续寻找差异基因、选择性剪切事件检测等分析结果。因此准确的计算出转录组表达水平对转录组研究具有重要的作用。本文在先前提出的用于计算外显子芯片基因及异构体表达水平的GME模型基础上,提出了i GME模型,利用已知基因、剪接异构体、探针的对应关系,模拟了条件独立的探针特性,加速原来GME的计算速度并获得了较为准确的精度。针对i GME模型单条件的芯片个数增加,模型效率降低的问题,提出一种充分利用多核处理机或者集群环境来提高效率的并行计算方法。通过实验验证并行计算极大地提高了模型的计算效率,与先前的串行计算相比,并行计算使得i GME模型更适用于大规模的外显子芯片分析。最后,通过构造新的基因、异构体、探针的映射关系将GME模型用于Affymetrix新设计的HTA2.0芯片基因与异构体表达水平的计算。实验证明GME模型能够较为准确地计算出HTA2.0芯片的基因和异构体表达水平。
[Abstract]:Studies have shown that selective splicing is common in higher eukaryotes. Abnormal splicing may lead to abnormal transcriptome, such as Parkinson's disease, Alzheimer's disease, myotonic dystrophy and so on. Therefore, the study of transcriptome expression is a hot topic in biomedical field in recent years. The analysis of gene and isomer expression level based on gene microarray provides a feasible way to study the changes of transcriptional expression. Exon chip and HTA2.0 chip prepared by Affmetrix Company are involved in transcriptome research. Using the gene chip technology. In the transcriptome study, the accuracy of the level of transcriptome expression affects the results of subsequent search for differentially expressed genes, selective shear event detection and so on. Therefore, accurate calculation of transcriptome expression level plays an important role in transcriptome research. Based on the GME model used to calculate the expression level of exon microarray genes and isomers, I GME model is proposed in this paper. Using the corresponding relationships of known genes, splicing isomers and probes, the conditional independent probe properties are simulated. The calculation speed of the original GME is accelerated and the accuracy is obtained. In order to solve the problem of increasing the number of single chips and reducing the efficiency of the model, a parallel computing method is proposed to improve the efficiency of I GME model by making full use of multi-core processors or cluster environments. It is proved by experiments that the parallel computation greatly improves the computational efficiency of the model. Compared with the previous serial computation, the parallel computing makes the iGME model more suitable for large-scale exon chip analysis. Finally, the GME model is used to calculate the expression level of gene and isomer in the newly designed Affymetrix chip by constructing the mapping relationship of new genes, isomers and probes. The experimental results show that the GME model can accurately calculate the gene and isomer expression levels of HTA2.0 chip.
【学位授予单位】：南京航空航天大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：Q78;TP338.6

【相似文献】