凝膠圖像間的蛋白點(diǎn)匹配方法研究
發(fā)布時(shí)間：2019-04-22 11:45
【摘要】：人類基因組全序列測(cè)定的完成標(biāo)志著后基因時(shí)代的到來，生命科學(xué)從基因組序列分析轉(zhuǎn)向基因功能的研究。蛋白質(zhì)組學(xué)是后基因時(shí)代研究中的重要組成部分，雙向凝膠圖像分析技術(shù)是蛋白質(zhì)組學(xué)研究的重要技術(shù)，差異蛋白點(diǎn)提取和分析是雙向凝膠圖像分析技術(shù)的核心內(nèi)容，蛋白點(diǎn)匹配又是差異蛋白點(diǎn)提取的關(guān)鍵環(huán)節(jié)。蛋白點(diǎn)匹配算法可用于鑒定潛在差異蛋白質(zhì)分子，能夠?yàn)榧膊≡\斷、藥物研制和環(huán)境污染分析提供依據(jù)。 本論文以凝膠電泳圖像蛋白點(diǎn)為研究對(duì)象，探索凝膠圖像間的蛋白點(diǎn)匹配方法。主要研究工作、成果和創(chuàng)新點(diǎn)如下： 首先，介紹了蛋白組學(xué)的發(fā)展和主要關(guān)鍵技術(shù)，概述了本課題的研究背景意義，綜述了凝膠圖像蛋白點(diǎn)匹配方法。 然后，通過添加Landmark標(biāo)記點(diǎn)和運(yùn)用近鄰點(diǎn)機(jī)制，提出了基于Landmark的兩凝膠圖像間的蛋白點(diǎn)半自動(dòng)匹配算法。算法通過采用Landmark區(qū)域劃分法和對(duì)應(yīng)區(qū)域和跨區(qū)域蛋白點(diǎn)匹配提高了算法精度，并從Landmark數(shù)目和圖源兩角度驗(yàn)證了算法的有效性。但此方法為半自動(dòng)方法，，需要人工介入。 其次，依據(jù)凝膠電泳圖像蛋白點(diǎn)的分布特征，提出了先特征蛋白點(diǎn)的粗匹配再兩圖像間蛋白點(diǎn)精匹配的自動(dòng)匹配算法。首先采用統(tǒng)計(jì)學(xué)原理計(jì)算匹配距離閾值；然后采用區(qū)域劃分和相似性原理進(jìn)行特征點(diǎn)提取及粗匹配；其次利用匹配到的特征點(diǎn)建立兩圖像間映射關(guān)系；最后實(shí)現(xiàn)兩圖像蛋白點(diǎn)間的精確匹配。通過真實(shí)凝膠圖像驗(yàn)證了算法的有效性。 最后，從醫(yī)學(xué)分析效果角度，采用了先組內(nèi)兩兩匹配并生成合成膠，再合成膠蛋白點(diǎn)匹配的多幅凝膠圖像間蛋白點(diǎn)匹配策略。將凝膠圖像間的蛋白點(diǎn)匹配方法嵌入到凝膠圖像蛋白點(diǎn)分析軟件Protein Master中，并對(duì)Protein Master軟件匹配系統(tǒng)中的各子模塊進(jìn)行了簡(jiǎn)單介紹。
[Abstract]:The completion of complete sequencing of the human genome marks the arrival of the post-gene era, and life science has shifted from genomic sequence analysis to gene function research. Proteomics is an important part of post-gene age research. Bi-directional gel image analysis is an important technology in proteomics research. Differential protein spot extraction and analysis is the core of bi-directional gel image analysis. Protein dot matching is also the key to differential protein spot extraction. Protein dot matching algorithm can be used to identify potentially differential protein molecules, which can provide basis for disease diagnosis, drug development and environmental pollution analysis. In this paper, the protein spots of gel electrophoresis images are studied, and the matching methods of protein spots between gel images are explored. The main research work, achievements and innovations are as follows: firstly, the development and key technologies of proteomics are introduced, the background significance of this subject is summarized, and the methods of protein dot matching in gel image are summarized. Then, a semi-automatic protein point matching algorithm between two gel images based on Landmark is proposed by adding Landmark labeling points and applying the nearest neighbor point mechanism. The algorithm improves the accuracy of the algorithm by using the Landmark region partition method and the matching of the corresponding and cross-region protein spots. The effectiveness of the algorithm is verified from the point of view of the number of Landmark and the source of the map. However, this method is semi-automatic and requires manual intervention. Secondly, according to the distribution characteristics of protein spots in gel electrophoresis images, an automatic matching algorithm based on coarse matching of feature protein spots and fine matching of protein spots between two images is proposed. Firstly, the matching distance threshold is calculated by using the statistical principle; then the feature points are extracted and coarse matching is carried out by using the region division and similarity principle; secondly, the mapping relationship between the two images is established by using the matching feature points. Finally, the exact matching between the two image protein spots is realized. The validity of the algorithm is verified by real gel images. Finally, from the point of view of the effect of medical analysis, a multi-image matching strategy of protein spots between gel images was adopted, in which pairwise matching was first used to generate synthetic glue and then synthetic gelatinous spot matching. The protein point matching method between gel images is embedded into the gel image protein dot analysis software Protein Master and the sub-modules of the Protein Master software matching system are briefly introduced.
【學(xué)位授予單位】：南昌航空大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2011
【分類號(hào)】：R346
[Abstract]:The completion of complete sequencing of the human genome marks the arrival of the post-gene era, and life science has shifted from genomic sequence analysis to gene function research. Proteomics is an important part of post-gene age research. Bi-directional gel image analysis is an important technology in proteomics research. Differential protein spot extraction and analysis is the core of bi-directional gel image analysis. Protein dot matching is also the key to differential protein spot extraction. Protein dot matching algorithm can be used to identify potentially differential protein molecules, which can provide basis for disease diagnosis, drug development and environmental pollution analysis. In this paper, the protein spots of gel electrophoresis images are studied, and the matching methods of protein spots between gel images are explored. The main research work, achievements and innovations are as follows: firstly, the development and key technologies of proteomics are introduced, the background significance of this subject is summarized, and the methods of protein dot matching in gel image are summarized. Then, a semi-automatic protein point matching algorithm between two gel images based on Landmark is proposed by adding Landmark labeling points and applying the nearest neighbor point mechanism. The algorithm improves the accuracy of the algorithm by using the Landmark region partition method and the matching of the corresponding and cross-region protein spots. The effectiveness of the algorithm is verified from the point of view of the number of Landmark and the source of the map. However, this method is semi-automatic and requires manual intervention. Secondly, according to the distribution characteristics of protein spots in gel electrophoresis images, an automatic matching algorithm based on coarse matching of feature protein spots and fine matching of protein spots between two images is proposed. Firstly, the matching distance threshold is calculated by using the statistical principle; then the feature points are extracted and coarse matching is carried out by using the region division and similarity principle; secondly, the mapping relationship between the two images is established by using the matching feature points. Finally, the exact matching between the two image protein spots is realized. The validity of the algorithm is verified by real gel images. Finally, from the point of view of the effect of medical analysis, a multi-image matching strategy of protein spots between gel images was adopted, in which pairwise matching was first used to generate synthetic glue and then synthetic gelatinous spot matching. The protein point matching method between gel images is embedded into the gel image protein dot analysis software Protein Master and the sub-modules of the Protein Master software matching system are briefly introduced.
【學(xué)位授予單位】：南昌航空大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2011
【分類號(hào)】：R346
【參考文獻(xiàn)】
相關(guān)期刊論文 前5條
1 安如,王慧麟,徐大新,馮學(xué)智,周紹光,何凱;基于影像尺度空間表達(dá)與魯棒Hausdorff距離的快速角點(diǎn)特征匹配方法[J];測(cè)繪學(xué)報(bào);2005年02期
2 劉珂;周富強(qiáng);張廣軍;;半徑約束最小二乘圓擬合方法及其誤差分析[J];光電子·激光;2006年05期
3 汪亞明;圖像匹配的魯棒型Hausdorff方法[J];計(jì)算機(jī)輔助設(shè)計(jì)與圖形學(xué)學(xué)報(bào);2002年03期
4 李伯良,李林,吳家睿;功能蛋白質(zhì)組學(xué)[J];生物工程進(jìn)展;1999年04期
5 李成;黃歡;孫嘯;周國(guó)華;;凝膠微球芯片實(shí)驗(yàn)數(shù)據(jù)分析方法的設(shè)計(jì)與應(yīng)用[J];中國(guó)生物醫(yī)學(xué)工程學(xué)報(bào);2010年04期
相關(guān)博士學(xué)位論文 前1條
1
本文編號(hào)：2462806
本文鏈接：http://sikaile.net/xiyixuelunwen/2462806.html
最近更新
教材專著
