2011年11月24日
Prediction of protein residue contacts using discriminative random field
研究報告数理モデル化と問題解決(MPS)
- ,
- ,
- ,
- 巻
- 2011
- 号
- 13
- 開始ページ
- 1
- 終了ページ
- 6
- 記述言語
- 英語
- 掲載種別
Understanding interaction between proteins provides a clue to the mechanisms of protein function. Protein residues at interacting sites have co-evolved with those at the corresponding residues in the partner protein to keep their interactions. Therefore, mutual information between residues calculated from multiple sequence alignments of homologous proteins is considered to be useful for identifying contact residues in interacting proteins. The discriminative random field (DRF) is a special type of conditional random fields and can recognize some specific characteristic regions in an image. Since the matrix consisted of correlation between residues can be regarded as an image, we propose a prediction method for protein residue contacts using DRF models with correlation scores between residues based on mutual information. In this work, we perform computational experiments for several interactions between Pfam domains and discuss the results.Understanding interaction between proteins provides a clue to the mechanisms of protein function. Protein residues at interacting sites have co-evolved with those at the corresponding residues in the partner protein to keep their interactions. Therefore, mutual information between residues calculated from multiple sequence alignments of homologous proteins is considered to be useful for identifying contact residues in interacting proteins. The discriminative random field (DRF) is a special type of conditional random fields and can recognize some specific characteristic regions in an image. Since the matrix consisted of correlation between residues can be regarded as an image, we propose a prediction method for protein residue contacts using DRF models with correlation scores between residues based on mutual information. In this work, we perform computational experiments for several interactions between Pfam domains and discuss the results.
- リンク情報
-
- CiNii Articles
- http://ci.nii.ac.jp/naid/110008694716
- CiNii Books
- http://ci.nii.ac.jp/ncid/AN10505667
- URL
- http://id.nii.ac.jp/1001/00078730/
- ID情報
-
- CiNii Articles ID : 110008694716
- CiNii Books ID : AN10505667