基于蛋白质哈斯矩阵图的分泌蛋白预测

Predicting secretory proteins based on protein Hasse matrix image

因为研究分泌蛋白质有助于找到直接与特定生理或病理状态相关的生物分子,判断一条未知蛋白是否为分泌蛋白是非常重要的。基于同一类型蛋白质的哈斯矩阵图具有相似图像纹理假设,提取图像的几何矩作为伪氨基酸成分对未知蛋白质序列是否属于分泌蛋白进行预测,采用Jackknife算法进行测试,预测成功率与现有算法相比有很大的提高。

It is important to identify whether an uncharacterized protein sequence is a secretory proteins or not because secretory proteins are composed with signal peptides which are crucial tool in finding new drugs or reprogramming cells for gene therapy.Based on the assumption that proteins belonging to a same class must bear some sort of similar texture on the protein Hasse matrix images, geometric invariant moment factors derived from the image are used as the pseudo amino acid components to formulate the protein samples for statistical prediction.The success rates obtained on a previously constructed benchmark dataset are quite promising.