High-accuracy protein structure modeling demands accurate and very fast side chain prediction since such a procedure must be repeatedly called at each step of structure refinement. Many known side chain prediction programs, such as SCWRL and TreePack, depend on the philosophy that global information and pairwise energy function must be used to achieve high accuracy. These programs are too slow to be used in the case when side chain packing has to be used thousands of times, such as protein structure refinement and protein design. We present an unexpected study that local backbone information can determine side chain conformations accurately. LocalPack, our side chain packing program which is based on only local information, achieves equal accuracy as SCWRL and TreePack, yet runs 4-14 times faster, hence providing a key missing piece in our efforts to high-accuracy protein structure modeling.
|Title of host publication
|Research in Computational Molecular Biology - 12th Annual International Conference, RECOMB 2008, Proceedings
|Number of pages
|Published - 2008
|"12th Annual InternationalConference on REsearch in COmputational Molecular Biology, RECOMB 2008" - Singapore, Singapore
Duration: Mar 30 2008 → Apr 2 2008
|Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
|"12th Annual InternationalConference on REsearch in COmputational Molecular Biology, RECOMB 2008"
|03/30/08 → 04/2/08
- Local backbone features
- Multiclass Support vector machines
- Side chain prediction
ASJC Scopus subject areas
- Theoretical Computer Science
- General Computer Science