山东大学学报(工学版)
山東大學學報(工學版)
산동대학학보(공학판)
JOURNAL OF SHANDONG UNIVERSITY(ENGINEERING SCIENCE)
2014年
2期
12-20
,共9页
魏小敏%徐彬%关佶红
魏小敏%徐彬%關佶紅
위소민%서빈%관길홍
蛋白质相互作用%能量热点%特征选择%递归消除%预测
蛋白質相互作用%能量熱點%特徵選擇%遞歸消除%預測
단백질상호작용%능량열점%특정선택%체귀소제%예측
protein-protein interaction%energy hot spots%feature selection%recursion eliminate%prediction
基于蛋白质相互作用能量热点的特性,定义了残基接触数、溶剂可及性面积相对变化量所占比例等18个新特征。分别使用基于支持向量机(support vector machine,SVM)和基于 F - Score 的递归特征消除法进行特征选择,提出对应的预测模型 SVM - RFE 和 F - Score - RFE 用于蛋白质能量热点的预测。实验结果显示,在独立测试中 F - Score - RFE 模型的 F1比当前预测性能最好的方法提高6.25%,表明所定义的新特征对蛋白质能量热点的识别具有较大的贡献。
基于蛋白質相互作用能量熱點的特性,定義瞭殘基接觸數、溶劑可及性麵積相對變化量所佔比例等18箇新特徵。分彆使用基于支持嚮量機(support vector machine,SVM)和基于 F - Score 的遞歸特徵消除法進行特徵選擇,提齣對應的預測模型 SVM - RFE 和 F - Score - RFE 用于蛋白質能量熱點的預測。實驗結果顯示,在獨立測試中 F - Score - RFE 模型的 F1比噹前預測性能最好的方法提高6.25%,錶明所定義的新特徵對蛋白質能量熱點的識彆具有較大的貢獻。
기우단백질상호작용능량열점적특성,정의료잔기접촉수、용제가급성면적상대변화량소점비례등18개신특정。분별사용기우지지향량궤(support vector machine,SVM)화기우 F - Score 적체귀특정소제법진행특정선택,제출대응적예측모형 SVM - RFE 화 F - Score - RFE 용우단백질능량열점적예측。실험결과현시,재독립측시중 F - Score - RFE 모형적 F1비당전예측성능최호적방법제고6.25%,표명소정의적신특정대단백질능량열점적식별구유교대적공헌。
18 new features such as residue contact number and the proportion of relative change of accessible surface area et al. were derived based on the analysis of protein-protein interaction energy hot spots. Two recursion feature elimina-tion methods were used to select discriminative feature subsets and two corresponding prediction models were proposed, noted as SVM - RFE and F - Score - RFE. The experimental results showed that the prediction model F - Score - RFE could improve 6. 25% in the value of F1 compared with the best existing method on the same independent test dataset, which indicated that new features defined were significant to improve the performance of prediction.