Premium
ScreenCap3: Improving prediction of caspase‐3 cleavage sites using experimentally verified noncleavage sites
Author(s) -
Fu SzuChin,
Imai Kenichiro,
Sawasaki Tatsuya,
Tomii Kentaro
Publication year - 2014
Publication title -
proteomics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.26
H-Index - 167
eISSN - 1615-9861
pISSN - 1615-9853
DOI - 10.1002/pmic.201400002
Subject(s) - cleavage (geology) , caspase , caspase 3 , apoptosis , computational biology , chemistry , computer science , biology , biochemistry , programmed cell death , paleontology , fracture (geology)
Because of its wide range of substrates, caspase-3, a main executioner among apoptosis-related caspases, is thought to have many unknown substrates that have remained unidentified. This report describes our predictive method to facilitate the discovery of novel caspase-3 substrates. To develop a more reliable prediction method, we specifically examined improvement of the data quantity and quality of caspase-3 cleavage sites. The ScreenCap3 method is based on machine learning and on information not only of experimentally verified positive examples but also of negative examples, which were not cleaved by caspase-3. Using information of experimentally verified noncleavage sites, we elucidate novel patterns of amino acids around "actual" cleavage sites. Results show that ScreenCap3 provides substantial improvement in terms of precision, compared with existing methods. Therefore, ScreenCap3 is anticipated for use with proteomic screening and identification of novel caspase-3 substrates and their cleavage sites. ScreenCap3 is available at http://scap.cbrc.jp/ScreenCap3/.