蛋白质家族和结构域DOC

Pfam

（

Protein families database of alignments and HMMs

）

A database with protein domain

derived from sequences in SWISSPROT and TrEMBL. Each motif or domain is represented by an

HMM

profile

generated

from

the

seed

alignment

of

a

number

of

conserved

homologous

proteins. /

The Pfam database is composed of two parts

Pfam-A involves manual alignments

Pfam-B, automatic alignment in a way similar to ProDom

（

PSI-BLAST

）

.

The functional annotation of motifs in Pfam-A is often related to that in PROSITE. Pfam-B only

contains sequence families not covered in Pfam-A.

Because of the automatic nature, Pfam-B has a much larger coverage but is also more error prone

because some HMMs are generated from unrelated sequences.

SMART (Simple Modular Architecture Research Tool

）：

Contains

HMM

profiles

constructed

from

manually

refined

protein

domain

alignments. /

Alignments in the database are built based on

tertiary structures whenever available

or based on PSI-BLAST profiles.

Alignments

are

further

checked

and

refined

by

human

annotators

before

HMM

profile

construction.

Protein functions are also manually curated.

The database may be of better quality than Pfam with more extensive functional annotations.

Compared

to

Pfam,

the

SMART

database

contains

an

independent

collection

of

HMMs,

with

emphasis on signaling, extracellular, and chromatin-associated motifs and domains.

Sequence searching in this database produces a graphical output of domains with well- annotated

information

with

respect

to

cellular

localization,

functional

sites,

superfamily,

and

tertiary

structure

InterPro

：

An integrated pattern database /interpro/

The

database

integrates

information

from

PROSITE,

Pfam,

PRINTS,

ProDom,

and

SMART

databases.

The sequence patterns from the five databases are further processed. Only overlapping motifs and

domains in a protein sequence derived by all five databases are included.

A popular feature of this database is a graphical output that summarizes

motif

matches and has

links to more detailed information.

CDD( Conserved Domain Database)

a

collection

of

multiple

sequence

alignments

for

ancient

domains

and

full-length

proteins. /Structure/cdd/

The CD-Search service may be used to identify the conserved domains present in a protein query

sequence: /Structure/cdd/

RPS-BLAST (Reverse PSI- BLAST) is the search tool used in the CD-Search service.

uses a query sequence to search against a pre-computed profile database generated by PSI-BLAST.

The

role

of

the

PSSM

has

changed

from

to

hence

the

term

in

RPS-BLAST.

It

performs

only

one

iteration

of

regular

BLAST

searching

against

a

database

of

PSI-BLAST

profiles to find the high-scoring gapped matches.

CDART (Conserved Domain Architecture) :

A domain search program /BLAST/

Combines the results from RPS- BLAST, SMART, and Pfam.

The

resulting

domain

architecture

of

a

query

sequence can

be

graphically

presented

along

with

related sequences.

CDART is not a substitute for individual database searches because it often misses certain features

that can be found in SMART and Pfam.

1.2

蛋白质家族数据库

COG (Cluster of Orthologous Groups ):

A

protein

family

database

based

on

phylogenetic

classification. /COG/

It is constructed by comparing protein sequences encoded in completely sequenced genomes.

Unicellular clusters

：检索工具为

COGnitor program

Eukaryotic Clusters

：检索工具为

KOGnitor

A

query

sequence

can

be

assigned

function

if

it

has

significant

similarity

matches

with

any

member of the cluster.

ProtoNet:

A

database

of

clusters

of

homologous

proteins

similar

to

COG

. /

Orthologous

protein

sequences

in

the

SWISSPROT

database

are

clustered

based

on

pairwise

sequence comparisons between all possible protein pairs using BLAST.

Protein relatedness is defined by the E-values from the BLAST alignments.

A query protein sequence can be submitted to the server for cluster identification and functional

annotation.

1.3

、蛋白质结构数据库

PDB

（

Protein Data Bank

）

PDB

< br>中含有通过实验（

X

射线晶体衍射，核磁共振

）测定的生物大分子的三维结构

蛋白质

核酸

糖类

其它复合物

/pdb

SCOP

（

Structural Classification of Proteins

）蛋白质结构分类数据库

提供关于已 知结构的蛋白质之间结构和进化关系的详细描述，包括蛋白质结构数据库

PDB

中的所有条目。

/scop/

SCOP

数据库除了提供蛋白质结构和进化关系 信息外，对于每一个蛋白质还包括下述信息：

到

PDB

的连接，序列，参考文献，结构的图像等。

可 以按结构和进化关系对蛋白质分类，

分类结果是一个具有层次结构的树，

其主要的层次是

家族、超家族和折叠

:

家族：具有明显的进化关系

超家族：具有远源进化关系，具有共同的进化源

折叠类：主要结构相似

DSSP

（蛋白质二级结构数据库）

对生物大分子数据库

PDB

中的任何一 个蛋白质，根据其三维结构推导出对应的二级结构。

/dssp/

对研究蛋白质序列与蛋白质二级结构及空间结构的关系非常有用

除了二级结构以外，

DSSP

还包括蛋 白质的几何特征及溶剂。

HSS P

（蛋白质同源序列比对数据库）

二级数据库

/hssp/

数据来源于

PDB

，或 来源于

SWISS-PROT

对于

P DB

中的每一个蛋白质，

HSSP

将与 其同源的所有蛋白质序列对比排列起来，从而将

相似序列的蛋白质聚集成结构同源的家族 。

HSSP

有助于分析蛋白质的保守 区域，研究蛋白质的进化关系，有助于蛋白质的分子设计。

1.4

、其它生物大分子数据库

MMDB

（

Molecular Modeling Database

）

MMDB

是

（

NCBI

）

Entrez

的一个部 分，

数据库的内容包括来自于实验的生物大分子结构数

据。

/entrez/?db=Structure

与

PDB

相比，对于数据库中的每一个生物大分子结构，

MMDB

具有许多附加的信息，如分

子的生物学功能、产生功能 的机制、分子的进化历史等

。

还提供生物大分子三维结构模型显示、结构分析和结构比较工具。

dbSNP

（

Single nucleotide polymorphisms

，单核苷酸多态性数据库）

/entrez/?db=snp

OMIM (Online Mendelian Inheritance in Man)

是关于人类基因和遗传疾病的分类数据库