添加二级结构作为特征来对抗癌肽进行预测

Addition of Protein Secondary Structure Information for Prediction of Anticancer Peptide

摘要: 抗癌肽是一种具有明显抗肿瘤活性的抗微生物肽，抗癌肽不仅能快速高效地消灭致病病菌，还能有效地作用于人体肿瘤细胞。本文将已发表文献的抗癌肽数据集中，首次提取了蛋白质3种二级结构组分(3PSS)作为特征参量，并结合20种氨基酸组分(20AAC)和6种亲疏水氨基酸组分(6HP)作为特征信息，并采用二次判别法(QD)实施预测。最后，在7折交叉检验下，当采用蛋白质3种二级结构组分(3PSS)结合6种亲疏水氨基酸组分(6HP)作为特征时，预测总精度(Acc)达到86%；当采用蛋白质3种二级结构组分(3PSS)结合20种氨基酸组分(20AAC)作为特征时，预测总精度达到94%。从预测结果发现：添加了二级结构信息后，预测精度都有不同程度的提高。另外，在同种数据集中，和其他预测软件相比较，再次确认了加入二级结构信息作为特征后，我们的预测精度是最高的。

Abstract: The anticancer peptide is a kind of antimicrobial peptide which has obvious antitumor activity. Anti-cancer peptide can not only quickly and effectively eliminate pathogenic bacteria, but also can be effective in human tumor cells. Based on the published literature anticancer peptide data-set, the three kinds of protein secondary structure (3PSS) were extracted as the characteristic parameters for the first time. Combined with 20 kinds of amino acids (20AAC) and 6 kinds of hydrophobic amino acids (6HP) as characteristic information, using the quadratic discriminant method (QD) to carry out prediction, under the 7 folds of cross-validations, when using three kinds of protein secondary structure components (3PSS) combined with six kinds of hydrophobic amino acid components (6HP) as a feature, the overall accuracy (Acc) was 86%. When using three kinds of protein secondary structure components (3PSS) combined with 20 kinds of amino acid components (20AAC) as a feature, the total accuracy was 94%. It is found from the prediction results that the total accuracy can be improved with the addition of secondary structure information. In addition, compared with other prediction software, we confirm that the prediction accuracy was the highest when we join the secondary structure.