東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Mechanistic models of protein evolut...

Dimmic, Matthew Wayne.

FindBook

Google Book

Amazon

博客來

Mechanistic models of protein evolution for phylogenetic inference.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Mechanistic models of protein evolution for phylogenetic inference./
作者:	Dimmic, Matthew Wayne.
面頁冊數:	150 p.
附註:	Source: Dissertation Abstracts International, Volume: 64-02, Section: B, page: 0579.
Contained By:	Dissertation Abstracts International64-02B.
標題:	Biophysics, General. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3079433

Mechanistic models of protein evolution for phylogenetic inference.
Dimmic, Matthew Wayne.

Mechanistic models of protein evolution for phylogenetic inference. - 150 p.

Source: Dissertation Abstracts International, Volume: 64-02, Section: B, page: 0579.

Thesis (Ph.D.)--University of Michigan, 2003.

Proteins play a vital role in almost every process of life, and this diversity has been achieved by the selective forces of evolution. By developing a fundamental understanding of the factors important to protein evolution, we can learn a great deal about the natural laws governing protein structure and function. To date, computational evolutionary models for amino acid sequences are relatively rudimentary. They typically provide fixed, non-mechanistic parameters which yield little information about specific evolutionary processes, and the variety of chemical environments in the folded protein is ignored.Subjects--Topical Terms:

1019105
Biophysics, General.

Mechanistic models of protein evolution for phylogenetic inference.
LDR:03375nmm 2200325 4500 001 1858941
005 20041020095006.5
008 130614s2003 eng d
035 $a (UnM)AAI3079433
035 $a AAI3079433
040 $a UnM $c UnM
100 1 $a Dimmic, Matthew Wayne. $3 1946610
245 1 0 $a Mechanistic models of protein evolution for phylogenetic inference.
300 $a 150 p.
500 $a Source: Dissertation Abstracts International, Volume: 64-02, Section: B, page: 0579.
500 $a Chair: Richard A. Goldstein.
502 $a Thesis (Ph.D.)--University of Michigan, 2003.
520 $a Proteins play a vital role in almost every process of life, and this diversity has been achieved by the selective forces of evolution. By developing a fundamental understanding of the factors important to protein evolution, we can learn a great deal about the natural laws governing protein structure and function. To date, computational evolutionary models for amino acid sequences are relatively rudimentary. They typically provide fixed, non-mechanistic parameters which yield little information about specific evolutionary processes, and the variety of chemical environments in the folded protein is ignored.
520 $a We have implemented several models in a maximum likelihood framework which address these shortcomings. A model for reverse transcriptases (RT's) was developed, called rtREV. Optimized on retroviral sequences including HIV, it is found to be applicable to a diverse range of RT's both endogenous and exogenous, demonstrating that the model encapsulates evolutionary information specific to that protein family.
520 $a Mechanistic models called amino acid fitness models use hidden site classes to represent a variety of selective constraints on amino acid substitutions, yet make no prior assumptions about which locations in the protein are under which constraints. When applied to mitochondrial protein families, these models showed a significantly better fit to the data than site-homogeneous models. In studies on families of G protein-coupled receptors (GPCR's), fitness models were found to reproduce structural and functional characteristics of the proteins without prior input of these characteristics. Using Bayesian posterior mapping, fitness models can identify transmembrane helices, ligand-binding regions, and regions of surface accessibility.
520 $a We introduced a model called BRASI, which accounts for different evolutionary constraints along branches of a phylogenetic tree. Using this model we reconstructed the amino acid composition of life's ancient cellular ancestor and compared it to the composition in modern sequences. Biases in this reconstruction under standard models are attributed to their failure to account for site and lineage heterogeneity.
520 $a Because current phylogenetic software does not implement hidden site class models, specialized software was developed with customizability as a goal, so that other researchers could use these models as well as design and test their own phylogenetic models.
590 $a School code: 0127.
650 4 $a Biophysics, General. $3 1019105
650 4 $a Biology, Biostatistics. $3 1018416
650 4 $a Biology, Molecular. $3 1017719
690 $a 0786
690 $a 0308
690 $a 0307
710 2 0 $a University of Michigan. $3 777416
773 0 $t Dissertation Abstracts International $g 64-02B.
790 1 0 $a Goldstein, Richard A., $e advisor
790 $a 0127
791 $a Ph.D.
792 $a 2003
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3079433