File S1 - In Silico Approach for Predicting Toxicity of Peptides and Proteins

File containing all supporting information figures and tables. Figure S1: Sequence logos of (A) first ten residues of N-terminus and (B) last ten residues of C-terminus of toxic peptides (alternate dataset), where size of residue is proportional to its propensity. Figure S2: Maximum and minimum scoring residues at every position as observed in quantitative matrix (alternate dataset). Table S1: Performance of whole amino acid and dipeptide composition-based SVM model developed on alternate dataset. Table S2: Performance of Binary profile-based models developed on alternate dataset. Table S3: Performance of motif based prediction (on alternate dataset). Table S4: Performance of hybrid model developed on alternate dataset. (DOC)本文件包含所有补充信息图表与表格。图S1：（A）毒性肽（替代数据集）N端前十个残基与（B）C端后十个残基的序列Logo（Sequence Logo），其中残基的大小与其倾向性成正比。图S2：替代数据集的定量矩阵中各位置上得分最高与最低的残基。表S1：基于替代数据集构建的全氨基酸组成与二肽组成支持向量机（SVM）模型的性能表现。表S2：基于替代数据集构建的二进制特征谱（Binary profile）模型的性能表现。表S3：针对替代数据集的基序（motif）预测方法的性能表现。表S4：基于替代数据集构建的混合模型的性能表现。（DOC）