Selected article for: "negative dataset and positive dataset"

Author: Yujia Xiang; Quan Zou; Lilin Zhao
Title: VPTMdb: a viral post-translational modification database
  • Document date: 2020_4_2
  • ID: kl99afiu_18
    Snippet: All viral phosphorylation experimentally verified serine sites in our database were used as positive samples, and those not marked by any phosphorylation information on the same protein were considered nega-tive samples. As a result, we obtained 182 phosphorylated serine residues as well as 2148 non-phosphorylated residues. Phosphorylated sites from UniProtKB/Swiss-Prot were regarded as the independent dataset, and they included 93 positive serin.....
    Document: All viral phosphorylation experimentally verified serine sites in our database were used as positive samples, and those not marked by any phosphorylation information on the same protein were considered nega-tive samples. As a result, we obtained 182 phosphorylated serine residues as well as 2148 non-phosphorylated residues. Phosphorylated sites from UniProtKB/Swiss-Prot were regarded as the independent dataset, and they included 93 positive serine sites and 1878 negative serine sites. After using CD-HIT (identity set to 0.8) (Fu, et al., 2012) to remove redundant sequences, we obtained 129 positive sites and 1611 negative sites. The independent dataset contained 52 positive sites and 1072 negative sites ( Table 1) . These sequences were truncated to a 23-residue symmetrical window (-11 to 11) (Supplementary Data Table) .

    Search related documents:
    Co phrase search for related documents
    • negative site and positive site: 1, 2, 3, 4, 5, 6
    • phosphorylated serine residue and serine residue: 1, 2, 3