fgl.ncsu.edu/smeng/GoAnnotationMagnaporthegrisea.html. Sequence similarity-based GO annotation Step 1 Predicted proteins of Version 5 of the M. oryzae genome sequence were
downloaded from the Broad Institute at http://www.broad.mit.edu/annotation/genome/magnaporthe_grisea/MultiDownloads.html. GO-annotated proteins were downloaded from the Gene Ontology (GO) database at http://www.Geneontology.org/GO.downloads.database.shtml. These GO-annotated proteins were from about 50 organisms with published association with GO terms. Only three of the 50 organisms are fungi. They are Candida albicans, Saccharomyces cerevisiae, and Schizosaccharomyces pombe. Other organisms are from bacteria, plants, or animals etc. Proteins of these non-fungal organisms were retained to 10058-F4 in vitro increase the number of proteins with validated selleckchem functions available for matching to M. oryzae. Step 2 Possible ortholog pairs between GO proteins and predicted proteins from M. oryzae genome sequence Version 5 were Alvocidib estimated by searching for reciprocal
best hits using BLASTP (e-value < 10-3) [24]. Step 3 Significant alignment pairs with 80% or better coverage of both query and subject sequences, 10-20 or less BLASTP E-value, and 40% or higher of amino acid identity (pid) were manually reviewed. Step 4 The functions of significantly matched GO proteins were manually cross- validated using data from wet lab experiments, see more and the NCBI Conserved Domain Database (CDD) [25]. Step 5 If the functions suggested from different sources were consistent with each other, and with available M. oryzae data, the functions of the experimentally characterized, significantly matched GO proteins, were transferred to the M. oryzae proteins in our study, and given the evidence code ISS (Inferred from Sequence Similarity) [26, 27]. Step 5 The information was recorded into a gene association file following the format standard at http://www.geneontology.org/GO.format.annotation.shtml. Literature-based GO annotation Step 1 Literature at public
databases such as PubMed [a database of biomedical literature citations and abstracts at the National Center for Biotechnology Information (NCBI)] were searched using key words, including alternative species names for the organism such as Magnaporthe grisea and Pyricularia oryzae. Step 2 Relevant published papers were read and genes or gene products and their functions were identified. Step 3 Where necessary, gene IDs and sequences at public databases, such as the NCBI protein database were identified. Step 4 Based on the functions identified in the paper(s), appropriate GO terms were found using AmiGO, the GO-supported tool for searching and browsing the Gene Ontology database. Step 5 Evidence codes were assigned following the guide at http://www.geneontology.org/GO.evidence.shtml.