PDbRNAbs: A prediction database of RNA binding sites in amino acid sequence using SVM
PDbRNAbs (Prediction Database of RNA binding sites) is a computational database that integrates a wealth of information relevant to human protein-RNA binding. The data source (a nonredundant data set of approximately 20000 proteins pertaining to human) was extracted from PDB (Protein Data Bank), and the SVM model was constructed using data instances encoded with three sequence features, including the side chain pKa value, hydrophobicity index and molecular mass of an amino acid 2. When compared with previous studies, the SVM model appears to be more accurate, and RNA-binding residues can be predicted at 66.28% sensitivity and 69.84% specificity. PDbRNAbs presents an interactive interface for users to query information regarding RNA binding sites prediction in human protein, and also allows users to obtain amino acid sequence annotated with header, acession date, compound, author list,resolution,experiment type (if not X-ray) and crystal unit cell parameters from the CRYST1 record.
computational database SVM model protein RNA binding
Yiming Zhang Peng Jiang Zuhong Lu
State Key Laboratory of Bioelectronics,Department of Biological Science and Medical Engineering, Sou State Key Laboratory of Bioelectronics,Department of Biological Science and Medical Engineering, Sou
国际会议
The 5th International Forum on Post-genome Technologies(5IFPT)(第五届国际后基因组生命科学技术学术论坛)
苏州
英文
2007-09-10(万方平台首次上网日期,不代表论文的发表时间)