國立高雄大學圖資館 |

語系: 繁體中文

說明(常見問題)

圖資館首頁

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Computational combinatorial protein ...

University of Pennsylvania.

Computational combinatorial protein design: Sequence sampling and statistical design.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Computational combinatorial protein design: Sequence sampling and statistical design.
作者:	Yang, Xi.
面頁冊數:	196 p.
附註:	Source: Dissertation Abstracts International, Volume: 66-06, Section: B, page: 3153.
附註:	Supervisor: Jeffery G. Saven.
Contained By:	Dissertation Abstracts International66-06B.
標題:	Chemistry, Physical.
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3179841
ISBN:	0542201135

Computational combinatorial protein design: Sequence sampling and statistical design.
Yang, Xi.

Computational combinatorial protein design: Sequence sampling and statistical design. - 196 p.

Source: Dissertation Abstracts International, Volume: 66-06, Section: B, page: 3153.

Thesis (Ph.D.)--University of Pennsylvania, 2005.

In the first approach, we introduce a powerful algorithm, biased Monte Carlo with replica exchange method (BMCREM), to achieve good energetic sampling in systems that have a rough energy landscape. The results obtained from the BMCREM method have been compared to the classic Monte Carlo with simulated annealing method (MCSA) and classic Monte Carlo with replica exchange (MCREM), and it shows the BMCREM method has greatly increased the sampling efficiency of difficult energy landscape. In the second approach, a statistical computationally assisted design strategy (SCADS) has been developed to take advantage of the simplified energy landscape. The SCADS method neglects the energy fluctuations, where the pair correlation is approximately treated. This approximation leads to a dramatic enhancement of the sampling efficiency. The result comparisons between the BMCREM and SCADS methods illustrate the role of pair correlations in generating probability profiles, thus demonstrate the merits and weakness of the two different approaches. The BMCREM and SCADS method then are applied to design an oligomer crystal structure, identify the key residues in the DNA-protein interface of engrailed homeodomain (1HDD), redesign a 20-residue "Trp-cage" protein to achieve ultrafast folding and select mutations of a new DNA-binding protein SWIRM.

ISBN: 0542201135Subjects--Topical Terms:

226924
Chemistry, Physical.

Computational combinatorial protein design: Sequence sampling and statistical design.
LDR:02997nmm _2200265 _450 001 167423
005 20061005085933.5
008 090528s2005 eng d
020 $a 0542201135
035 $a 00198039
040 $a UnM $c UnM
100 0 $a Yang, Xi. $3 237571
245 1 0 $a Computational combinatorial protein design: Sequence sampling and statistical design.
300 $a 196 p.
500 $a Source: Dissertation Abstracts International, Volume: 66-06, Section: B, page: 3153.
500 $a Supervisor: Jeffery G. Saven.
502 $a Thesis (Ph.D.)--University of Pennsylvania, 2005.
520 # $a In the first approach, we introduce a powerful algorithm, biased Monte Carlo with replica exchange method (BMCREM), to achieve good energetic sampling in systems that have a rough energy landscape. The results obtained from the BMCREM method have been compared to the classic Monte Carlo with simulated annealing method (MCSA) and classic Monte Carlo with replica exchange (MCREM), and it shows the BMCREM method has greatly increased the sampling efficiency of difficult energy landscape. In the second approach, a statistical computationally assisted design strategy (SCADS) has been developed to take advantage of the simplified energy landscape. The SCADS method neglects the energy fluctuations, where the pair correlation is approximately treated. This approximation leads to a dramatic enhancement of the sampling efficiency. The result comparisons between the BMCREM and SCADS methods illustrate the role of pair correlations in generating probability profiles, thus demonstrate the merits and weakness of the two different approaches. The BMCREM and SCADS method then are applied to design an oligomer crystal structure, identify the key residues in the DNA-protein interface of engrailed homeodomain (1HDD), redesign a 20-residue "Trp-cage" protein to achieve ultrafast folding and select mutations of a new DNA-binding protein SWIRM.
520 # $a The roughness of energy landscape requires efficient algorithms to be developed in protein design. To date, there are two different approaches to tackle the energy minima. One is to develop powerful algorithms to explicitly sample the local minima of the energy landscape; the other strategy uses an approximation to simplify the energy landscape, thus facilitates sampling. The former one requires extensive computational resource and time, while the latter one achieves efficiency by scarifying certain accuracy. Speed or accuracy is a dilemma in the protein design and protein folding. My thesis work mainly focuses on studying the two different approaches in the protein design.
590 $a School code: 0175.
650 # 0 $a Chemistry, Physical. $3 226924
690 $a 0494
710 0 # $a University of Pennsylvania. $3 212781
773 0 # $g 66-06B. $t Dissertation Abstracts International
790 $a 0175
790 1 0 $a Saven, Jeffery G., $e advisor
791 $a Ph.D.
792 $a 2005
856 4 0 $u http://libsw.nuk.edu.tw:81/login?url=http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3179841 $z http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3179841