Proteome analysis with the aid of the information oftians aiptome analysis for cancer diagnosis

从蛋白质组分析数据中，我们成功地找到了4个基因的表达，这4个基因的表达与癌症的进展和复发率密切相关。然后，我们试图借助转录组分析的信息，通过蛋白质组分析（PMF法）来研究基因表达水平，但不能清楚地检测到基因表达的蛋白质。特别是在AFP（抗胎蛋白）的情况下，其在我们的研究中与癌症复发率最强相关并且非常常见地作为肝癌的标志物，蛋白质斑点位置与白蛋白几乎相同。癌组织中白蛋白含量很高，AFP蛋白斑点被白蛋白斑点完全覆盖，我们尝试用多克隆抗体检测AFP蛋白。我们确定了癌组织中AFP的存在，但AFP的含量很低，制备多克隆抗体的过程耗时数天，而且制备过程不容易。该分析系统基于分水岭变换和滚球法，可以分析蛋白质斑点体积和斑点图像在显影过程中的时间过程。该软件可以分析重叠的蛋白质点，并检测少量蛋白质点，因为后期的imega具有许多点重叠的能力。然后我们可以在凝胶中检测1000多个蛋白质点，并识别150个点的蛋白质。该技术在美国电子学会年会上获得了第三届海报奖，但是我们所开发的技术不能检测到非常少量的蛋白质，如AFP、信号蛋白等蛋白质点。我们将在抗体工程的帮助下继续改进蛋白质组分析蛋白。

From proteome analysis data, we succeeded to find the 4 gene expression which are strongly related with the cancer progression and recurrence ratio by using the neural net work ratio. Then we tried to study the expression level by proteome analysis (PMF method) with the aid of the information of the transcriptome analysis.However, the proteins of the gene expression could not be detected clearly. Especially in the case of AFP (Anti feto protein), which is the most strongly related with the cancer recurrence ration in out study and very common as a marker of the hepatic cancer, the protein spot location is almost same as that of albumin. The albumin content of cancer tissue is very high and the AFP protein spot is covered perfectly by the albumin spot.Then we tried to use the polyclonal antibody for the detection of the AFP protein. We make sure the existence of the AFP in the cancer tissue, but the content is very low and we consume several days and procedure for the preparation of polyclonal antibody is not easy.Next we developed the 2-DE analysis system for the improvement of the spot detection and the quantification. The analysis system is based on the Watershed transformantion, and Rolling ball method, and can be analyze the time course of the protein spot volume and spot image during the developing process. This software can analyze the overlapping protein spots, and detect the spot of small amount protein, because of the ability of the late time imega which has many spot overlapping. Then we can detect more than 1000 protein spots in a gel and identify the proteins of 150 spots. This developed technique are evaluated in the annual meeting of American Electro Society, and we got the 3rd Place Poster Award.However, our developed the technique could not be detect the protein spot of very small amount of protein, for example, AFP, signal proteins. We will continue the improvement of the proteome analysis proteins with the aid of antibody engineering.