2018 Fiscal Year Annual Research Report
Prediction and analysis of intrinsically disordered regions in membrane proteins
| Project/Area Number |
17F17050
|
|---|
| Research Institution | The University of Tokyo |
|---|
Principal Investigator |
清水 謙多郎 東京大学, 大学院農学生命科学研究科(農学部), 教授 (80178970)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
FANG CHUN 東京大学, 農学生命科学研究科, 外国人特別研究員
|
|---|
| Project Period (FY) |
2017-10-13 – 2020-03-31
|
| Keywords | 天然変性 / タンパク質 / 深層学習 |
|---|
| Outline of Annual Research Achievements |
In the second year, our research focused on detail of "adopting feature fusion and feature compression method for identifying IDRs and motifs in IDPs"; the main works were as follows: (1) Applied effective feature-encoding scheme to combine more predictive features into fewer dimensions for prediction, it includes: (a) Remove the redundant features and strengthen the predictive features to enhance the accuracy of prediction. Using the scaling skills to enhance the predictive features and weaken the noise features; (b) Adopt the image processing technology to preprocess the conserved features included in PSSM; (c) Modify PSSM to combine the detailed local conservation patterns of residues with the distribution of scores in PSSM for prediction. (2) Adopted the feature fusion method, rather than connecting all features in series to design the algorithms, it includes: (a) Firstly, all the physicochemical features were clustered; (b) Secondly, factors were calculated to represent each clustering; (c) Finally, all features (including the revised PSSM and factors calculated from physicochemical features) were fused and compressed by matrix operations to reduce the feature dimensions. (3) For MoRFs in IDPs, detailed analysis was carried out according to their different lengths, and the related algorithms was designed respectively for them. (4) Implemented the related web tools for publication.
|
| Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
According to the methods that we developed, the world top prediction accuracy was achieved. We published two papers on the achievement.
|
| Strategy for Future Research Activity |
In the third year, we improve the performance of our MoRFs prediction system by integrating the amino acid sequence features designed in the previous year such as the revised PSSM and factors calculated from physicochemical features. We construct the Web site for MoRFs prediction and publish a paper of our research in the last four months.
|
