ＲＮＡ結合タンパク質の新規予測法

Chem-Bio Informatics Journal

木村 誠¹  久原 哲²  竹原 英毅²  喜久�? 真悟¹ 

[1] 九州大学大学院農学研究院生物機能科学部門生物機能化学講座;九州大学大学院農学研究院遺伝子資源工学部�?遺伝子制御学講座

关键词: hidden Markov models;    隠れマルコフモデル;    OB-fold;    ＯＢフォールド;    protein function prediction;    タンパク質の機能予測;    Pyrococcus horikoshii;    Pyococcus horikoshii;    RNA binding protein;    ＲＮＡ結合タンパク質;   

DOI  :  10.1273/cbij.5.1

学科分类：生物化学/生物物理

来源: Chem-Bio Informatics Society

【 摘 要 】

References(46)We have developed an efficient prediction procedure for RNA binding proteins with oligosaccharide/oligonucleotide binding-fold (OB-fold). First, all pairwise superimpositions of 96 OB-fold structures included in Structural Classification of Proteins (SCOP) database classified them into four distinct groups on the basis of structural similarity. The proteins belonging to each group were divided into RNA binding proteins and non-RNA binding proteins. The structure-based sequence alignment of RNA binding proteins in each group were made to build profile hidden Markov models (HMMs). The reliability of HMMs thus obtained was first evaluated by the application to the PDB40 sequence dataset; RNA binding proteins with OB-fold classified into OB-fold in SCOP were selected, giving E-values less than 1.0. The next application of HMMs to sequence database of the hyperthermophilic archaeon Pyrococcus horikoshii OT3 detected several RNA binding proteins, including tRNA synthetases, initiation factor, transcriptional regulatory proteins, and ribosomal protein L10E as RNA binding proteins with OB-fold. These results suggested that HMM derived from this study has information about RNA binding proteins with OB-fold. The present analysis strongly suggested 4 hypothetical proteins in P. horikoshii to be RNA binding proteins with OB-fold. Furthermore, the application of the present model to the rice full-length cDNA sequence database suggested 14 hypothetical proteins to be RNA binding proteins with OB-fold. It is known that some of the motifs have no specific biological function alone but are part of larger structural and functional assembles. Thus, the present method would provide clues as to protein functions of unannotated proteins and also be useful for a target selection for structural genomics.

【 授权许可】

Unknown