Three dimensional shape comparison of flexible protein using the local-diameter descriptor
Yi Fang1 Yu-Shen Liu1 Karthik Ramani1,2 |
|
|
|
Abstract |
Techniques for inferring the functions of the protein by comparing the shape similarity have been receiving a lot of attention. Proteins are functional units and their shape flexibility occupies an essential role in various biological processes. Several shape descriptors have demonstrated the capability of protein shape comparison by treating them as rigid bodies, which may give rise to incorrect comparison of flexible protein shape. We introduce an efficient approach for comparing flexible protein shapes by adapting a local diameter (LD) descriptor.
The LD descriptor, developed recently to handle skeleton based shape deformations [1], is adapted in this work to capture the invariant properties of shape deformations caused by the motion of the protein backbone. Every sampled point on the protein surface is assigned a value which measures the diameter of the 3D shape in the neighborhood of that point. The LD descriptor is built in form of one dimensional histogram from the distribution of the diameter values. The histogram based shape representation reduces the shape comparison problem of the flexible protein to a simple distance calculation between 1D feature vectors. Experimental results indicate how LD descriptor accurately treats the protein shape deformation. In addition, we use the LD descriptor for protein shape retrieval and compare it to the effectiveness of conventional shape descriptors. A sensitivity-specificity plot shows that LD descriptor performs much better than the conventional shape descriptors, in terms of consistency over a family of proteins and discernibility across families of different proteins.
|
Downloads |
|
Examples |
Visualization of two conformations (PDB: 1bpd(left) and 2bpg(right) with shape deformation. The protein in figure(B) transforms into protein in figure(A) by a rotation of the top portion of protein. The red arrow points to the contact region between the top portion and the bottom rigid part, which means that the topology changes during this deformation
Comparison of signatures of local diameter descriptor and Euclidean distance(ED) [Osada02]. descriptor,a popular igid method. (A)The shape signature built by local diameter descriptor (B)The signature built by ED descriptor. The red and blue histogram represent two proteins shape signatures respectively. From the figure, we can see the ignatures for two proteins by local diameter descriptor are much more consistent than that of ED shape signature. Note that, LD stands for local diameter shape signature and ED stands for Euclidean distance shape signature
A screen shot from flexible proteins shape retrieval system using our local diameter algortihm. In this figure, the query protein's information including its name and Group ID is shown on the top-left, and the local diameter values for each boundary are describled by the different colors, the color bar can be found in our paper. The query example from Group 1 has 30 deformation shapes, and the search system returns the first 15 proteins from the same group, the next two pages can find other 15 proteins.
|
Acknowledgment |
The database of Macromolecular Movements (MolMovDB) used in this paper is provided by Mark B Gerstein and Samuel C Flores. This material is partly based upon work supported by the National Science Foundation under Grant IIS No. 0535156. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation. We also acknowledge partial support from the National Institute of Health (GM-075004). |
[1] Gal R, Shamir A, Cohen-Or D: Pose-Oblivious Shape Signature. IEEE Transcations on Visualization and Computer Graphics 2007, 13:261–271