精华区文章阅读

发信人: zjliu (秋天的萝卜), 信区: Chemistry
标  题: [转载] 一步一步学计算机辅助药物设计(zz)
发信站: BBS 哈工大紫丁香站 (Fri Sep 17 16:02:07 2004)

发信人: whglc (九头鸟*), 信区: Chemistry
标  题: [转载] 一步一步学计算机辅助药物设计(zz)
发信站: 南京大学小百合站 (Wed Sep 15 12:39:38 2004)

最近很多人在做结构模拟,转一篇今天整理文件夹发现的文章,希望对大家有帮助.

一步一步学计算机辅助药物设计
Qianfeng
斑竹自述：
我不是“高手”，对于计算机辅助药物设计来说，只是一个初学者，在这里我会把我所知

道的写下来与大家交流，希望能结交众多志同道合的朋友。这里要特别感谢biolover的al

chemy,biohity曾经给我的帮助，他们是我心目中真正的高手。

1. 做生物大分子结构的pc机配置
转自biolover  biohity的推荐配置

1. 二手主板Tyan(泰安) Tiger MPX S2466N-4M 加两个Athlon MP 2400+ (是用XP 1700+改

装的）。总共3900元
2. 二手野猫4110显卡：$110.00 + ￥160关税。（未用上，因AGP口不配）。改用Synergy

II显卡：250元
3. 二手DC390U3W的SCSI卡：800元
4. 36G日立全新硬盘：1100元
5. SGI 1600SW 原装、全新液晶monitor：$1200 ＋￥1500（关税）
6. 美基P4-450瓦电源：450元；月光机箱：350元
7. 内存 512M;
8. 系统：win2000及Linux（在不同的硬盘上）。RedHat 7.2。

SGI的机子也很容易在国内买到极好的二手。O2-5K-200Mhz大约2100元即可买到。Octane-

R12k-300Mhz则贵些。但ebay上只要$300～400即可。PC机是无法与这类机子相提并论的。

要知道，1999年时，上述Octane要卖两万美元啊！

1、如果你在上海，进入易趣网站www.eachnet.com，在电脑-->工作站或服务器的栏目下，

找ninjaa这个人即可。
2、如果你在北京，找http://www.qianye.y365.com/newpage5.html并电话联系即可，此人

有Octane的各种机子卖，但要价太贵。
3、美国ebay上较便宜，但每台进口的Octane要交400元人民币左右的关税。

2. 再看看相关的教材和期刊
2. 再看看相关的教材和期刊
(1)入门教材
biolover  Alchemy推荐

Hyperchem里有一本书，叫“Computational Chemistry”，个人觉得比较经典。以前是独

立出来的PDF，现在集成到27M/2700页的大个头里面了。
刘次全编过一本“量子生物学”，随后编了“结构生物学”等，呵呵，神人
BioMedCAChe的UserGuide是比较通俗的

(2)推荐使用软件
Hyperchem，ICM，BioMedCAChe

(3)相关的部分专业期刊
转自biolover  biotech_lqh发贴

DDT（drug discovery today)
http://www.drugdiscoverytoday.com/

PROTEINS-STRUCTURE FUNCTION AND GENETICS
http://journals.wiley.com/0887-3585/

BIOINFORMATICS
http://bioinformatics.oupjournals.org/
http://bioinformatics.oupjournals.org/

BIOCHIMICA ET BIOPHYSICA ACTA-PROTEIN STRUCTURE AND MOLECULAR ENZYMOLOGY
http://www.elsevier.com/locate/issn/01674838

JOURNAL OF MEDICINAL CHEMISTRY
http://pubs.acs.org/journals/jmcmar/index.html

JOURNAL OF COMPUTER-AIDED MOLECULAR DESIGN
http://www.kluweronline.com/issn/0920-654X/contents

JOURNAL OF CHEMICAL INFORMATION AND COMPUTER SCIENCES
http://pubs.acs.org/journals/jcisd8/index.html

JOURNAL OF MOLECULAR GRAPHICS & MODELLING
http://www.elsevier.nl/locate/inca/525012

JOURNAL OF BIOMOLECULAR STRUCTURE & DYNAMICS  http://www.jbsdonline.com/

3. 还得熟悉一下有哪些大公司
开发分子模拟软件的一些公司及其网址
Tsar 、CAChe、Chem-X、UniChem、AMBER 、AMSOL 、AutoDock Oxford Molecular公司 h

ttp://www.oxmol.com/
ttp://www.oxmol.com/

Cerius2 、InsightII、WebLab、Catalyst、FELIX、QUANTA Molecular Simulations Inc

. (MSI) 公司      http://www.msi.com/

SYBYL, CoMFA, Biopolymer, TRIAD, GeneFold, MOLCAD, Alchemy Tripos
公司 http://www.tripos.com/

ChemDraw, Chem3D,ChemFinder,CambridgeSoft (CS)公司http://www.camsoft.com/

Gaussian Gaussian, Inc. 公司     http://www.gaussian.com/

SPARTAN、TITAN Wavefunction, Inc公司    http://www.wavefun.com/

PCMODEL  Serena Software公司      http://www.serenasoft.com/

MOPAC Fujitsu Ltd.公司        http://www.fujitsu.com/

Conformer   Princeton Simulations公司      http://www.conformer.com/

4. 开讲了，先讲讲分子对接，这里主要是指有机分子和受体的对接
受体当然是蛋白，有晶体结构的话，先学学PDB的格式(略)
蛋白结构模拟全攻略
蛋白结构模拟全攻略
1 先熟悉一下几个蛋白结构的数据库
Protein Structure Databases. Comparison and Classification of Protein Structur

es.

The study, comparison and classification of the already known protein structur

es allows to extract information about sequence/family/structure relationships

that can be used latter for predictions.

PDB. The database of protein structures. http://www.rcsb.org/
PDBsum. Summaries and structural analysis of PDB files.
http://www.biochem.ucl.ac.uk/bsm/pdbsum/

OCA. Integration of PDB with other sources of data. http://oca.ebi.ac.uk/

SCOP. Human expert classification of protein structures.
http://scop.mrc-lmb.cam.ac.uk/scop/

FSSP(Dali). Automatic classification. http://www.ebi.ac.uk/dali/

CATH. Semi-automatic classification.
http://www.biochem.ucl.ac.uk/bsm/cath_new/index.html

2 对于从来没做过结构的新手，可以看这里
The structure prediction Meta Server
http://bioinfo.pl/meta/
The structure prediction Meta Server offers a gateway to many high quality fol

d recognition servers and provides and infrastructure and main interface to se

veral highly reliable consensus methods. The Meta Server represents the firs s

tep in the our fold prediction strategy.
这是个大杂烩，是不是用了它其他的都不用了？  ；）

3 如果想多了解几个基于web的结构预测网站，可以看看这些
Web-based Comparative Protein Modelling Resources
SwissModel
http://www.expasy.ch/swissmod/
SWISS-MODEL is an Automated Protein Modelling Server running at GlaxoSmithKlin

e in Geneva (Switzerland) and the Advanced Biomedical Computing Center (NCI, F

rederick MD).

CPHModels
http://www.cbs.dtu.dk/services/CPHmodels/
Protein structure prediction server using distance constraints.
Center for Biological Sequence Analysis. (The Technical University of Denmark,

Denmark).
Denmark).

SDSC1
http://cl.sdsc.edu/hm.html
San Diego Supercomputer Center Protein Structure Homology Modeling Server

3D-JIGSAW
http://www.bmm.icnet.uk/servers/3djigsaw/
3D-JIGSAW is an automated system to build three-dimensional models for protein

s based on homologues of known structure.

ESyPred3D
http://www.fundp.ac.be/urbm/bioinfo/esypred/
ESyPred3D is an Expert System for the Prediction of proteins 3D structures run

ning at the University of Namur (Belgium). The specificity of ESyPred3D is the

building of the target-template alignment by combining several alignment prog

rams.

Pcomb-Pcons
http://www.sbc.su.se/~arne/pcomb/
Pcomb-Pcons uses a combination of several sequence-profile and profile-sequenc

e searches. It produces both homology modelling models and alignments.

Web-based Threading Resources
3D-PSSM
http://www.sbg.bio.ic.ac.uk/~3dpssm/
Protein fold recognition using 1D and 3D sequence profiles coupled with second

ary structure and solvation potential information.

FUGUE
http://www-cryst.bioc.cam.ac.uk/~fugue/
The program FUGUE searches a sequence or sequence alignment against the librar

y of profiles. Environment-specific substitution tables were derived from the

structure-based alignments in the HOMSTRAD database. Each alignment in HOMSTRA

D was converted into a scoring template (profile) using the environment-specif

ic substitution tables and environment-dependent gap penalties.

123
http://123d.ncifcrf.gov/

Libellula
http://www.pdg.cnb.uam.es:8081/libellula.html

SAMt99
http://www.cse.ucsc.edu/research/compbio/HMM-apps/model-library-search.html
http://www.cse.ucsc.edu/research/compbio/HMM-apps/model-library-search.html

SAUSAGE
http://rsc.anu.edu.au/~arussell/TheSausageMachine.html
Sausage (Sequence-structure Alignment Using a Statistical Approach Guided by E

xperiment) is a protein threading program.

Superfamily
http://supfam.mrc-lmb.cam.ac.uk/SUPERFAMILY/index.html
Protein domain assignments to SCOP structural superfamilies using a hidden Mar

kov model library.

bioinbgu
http://www.cs.bgu.ac.il/~bioinbgu/form.html

4 如果想看看预测后的结果，可以用一下Swiss-PdbViewer
同源模拟之swiss-model
SWISS-MODEL
http://www.expasy.org/swissmod/SWISS-MODEL.html
SWISS-MODEL is a fully automated protein structure homology-modeling server, a

ccessible via the ExPASy web server, or from the program DeepView (Swiss Pdb-V

iewer). The purpose of this server is to make Protein Modelling accessible to

all biochemists and molecular biologists World Wide.
all biochemists and molecular biologists World Wide.

Swiss-PdbViewer
http://www.expasy.org/spdbv/mainpage.htm
a tool for viewing and manipulating protein structures and models (Macintosh,

PC, SGI and Linux).

非常容易上手，适合初学者使用，有详细的说明，以及邮件列表可以咨询

5 如果觉得想自己动手，模拟一个，可以试试modeller
同源模拟之modeller
MODELLER

据说是目前最好的同源模拟软件
http://salilab.org/modeller/

MODELLER is used for homology or comparative modeling of protein three-dimensi

onal structures (1). The user provides an alignment of a sequence to be modele

d with known related structures and MODELLER automatically calculates a model

containing all non-hydrogen atoms. MODELLER implements comparative protein str

ucture modeling by satisfaction of spatial restraints (2, 3), and can perform

many additional tasks, including de novo modeling of loops in protein structur

many additional tasks, including de novo modeling of loops in protein structur

es, optimization of various models of protein structure with respect to a flex

ibly defined objective function, multiple alignment of protein sequences and/o

r structures, clustering, searching of sequence databases, comparison of prote

in structures, etc. MODELLER is written in Fortran 90 and runs on the Pentium

PC's (Linux and Win XP), Apple Macintosh (OS X) and workstations from Silicon

Graphics (IRIX), Sun (Solaris), IBM (AIX), and DEC Alpha (OSF/1).

6 模拟完得评价一下,不过这有些难度,我是没搞懂,请哪位大侠指点指点
Model Evaluation.
Once a model has been generated, these tools allow to evaluate its quality.

EVA
http://maple.bioc.columbia.edu/eva/
EVA continously and automatically analyses protein structure prediction server

s in 'real time' Columbia & Rockefeller Univ, New York.

Biotech Validation Suite.
http://biotech.embl-heidelberg.de:8400/
Based on the tools implemented in the WhatIf program.

WhatCheck / WhatIf Gert
WhatCheck / WhatIf Gert
http://www.cmbi.kun.nl/whatif//

ProSA II.
http://www.came.sbg.ac.at/Services/prosa.html

小结:其实模拟软件有很多，用的不同算法，各有优缺点，还有更专业的，专业的一般都是

Unix下的，象我这样的初级用户是不用想了

5. 接着如何取得有机小分子的配体呢，请看
biolover上这样说
问：谁知道怎么搭小分子配体的结构,然后再和蛋白做docking? 是有现成的立体结构库可

用,还是手工搭建?
答：一般是ISIS draw画出已知结构，Hyperchem等优化，找搞化学的人问问

ISIS/Draw
ISIS/Draw
http://www.mdli.com/
免费的画小分子结构的软件

一个简单介绍
http://www.ch.cam.ac.uk/SGTL/MDL/ISISdraw.html

其实网有免费2D转3D的软件CORINA
http://www2.chemie.uni-erlangen.de/software/corina/index.html
它需要输入SMILES格式
http://www.daylight.com/dayhtml/smiles/

还可以看看以下的帖子
如何寻求化合物信息

Chemfinder
http://chemfinder.cambridgesoft.com/

有机所的数据库
中药与有效成分数据库,药物与天然产物数据库等等
http://202.127.145.69/database/sjk.htm

还有两个大公司的产品目录也不错
Sigma
http://www.sigmaaldrich.com/

Acros
http://www.acros.be/
还能得到一些化合物的3D结构
还能得到一些化合物的3D结构

中国天然产物数据库（CNPD）
中国天然产物数据库
http://www.neotrident.com/neotrident_def4_1.htm

CNPD数据库包括两方面内容, 一是库管理系统, 二是数据本身。
1. CNPD的数据库是利用MDL公司先进的化学信息管理系统ISIS/Base & Draw进行管理应用

的。使用者无需任何专业计算机知识, 不需学习任何编程语言, 即可利用ISIS/Base & D

raw对CNPD数据库自如地进行检索、浏览、维护。
使用者在使用ISIS/Base & Draw对CNPD数据库进行检索时, 还可以使用到一种非常有效并

诱人的新功能--结构式搜寻。化学结构式是化学家的通用语言, 当化学家利用结构式搜寻

功能对CNPD数据库进行检索时, 可以非常准确的得到他们所需的信息。
Windows 环境下运行的数据库管理软件ISIS/Base&Draw, 使CNPD易于掌握使用。

2. CNPD 目前共收集了2000余条中国天然产物的化学结构及性质信息, 主要包括以下几类

天然产物:

生物碱 / 萜类 / 甾体类 / 皂苷类 / 鞣
黄酮类 / 香豆类 / 蒽醌类 / 木酚素类 / 类酯

CNPD收集的中国现有的天然产物的信息, 主要包括以下几个方面内容:

·天然产物的二维分子结构及利用分子力学方法优化得到的三维分子结构。三维分子结构

与生物活性资料相结合, 可探索分子结构和活性间的相互关系。
·天然产物的名称、分子式、分子量、熔点、旋光度等重要物理性质。
·天然产物的CAS登录号, 为方便地关联其它化学信息资源提供了便捷的接口。
·一些天然产物的生物活性信息。
·天然产物的植物来源。
·原植物在中国传统医药中的应用。通过与其它信息相关联, 可导出一些新的信息。
·参考文献。(包括文献来源、作者、年代等详细信息)

CNPD的应用

　　CNPD可以对从事天然产物、有机化学、医药、农药、植物化学及生物化学等领域

的研究员提供有益的帮助, 如:

·可以系统的了解天然产物分离、提取和鉴定的研究方法。
·可以对天然产物的结构及生物活性信息进行定量或定性的分析研究, 以深入了解某类药

物的结构与活性间的关系, 为新药研究提供系统的和有价值的信息。
·由于天然产物的结构具有非常好的差异性, 因此天然产物数据库可以作为一个非常有效

地组合化学化合物数据库的数据源。
·CNPD让你充分了解中国天然产物资源
　　→首次将天然产物、生物活性数据、原植物来源及中药传统应用融合在一起。
　　→三维结构信息及其在线显示。
　　→三维结构信息及其在线显示。

CNPD的升级/补充

　　中国的科学家在对天然产物及中草药研究中积累了大量的信息资料, CNPD将通过每年

的版本升级来充实完善数据库, 包括新条目的补充以及条目内容的增加等。

最低系统基本要求:

·IBM微机或100%兼容机；
·80486或更高微处理器；
·VGA或更好显示器
·微软Windows兼容鼠标
·微软Windows 95/98/2000/NT
·8MB以上内存
·硬盘空间: ISIS/Draw 15MB； ISIS/Base 40 MB； CNPD 6MB以上

CNPD DEMO CD CONTENT

·500 compounds in all, including:
·Alkaloids (生物碱) 75
·Carbohydrates (碳水化合物) 11
·Coumarins (香豆素) 36
·Coumarins (香豆素) 36
·Flavonoids (黄酮类似物) 40
·Terpenoids (萜类) 75
·Steroids (类固醇) 40
·Lignans (木脂素类) 61
·Anthraquinone (蒽醌) 37
·Phenols (苯酚) 58
·Polyacetylenes (多炔) 19
·Saponins (皂角苷) 40
·Lactones (内酯) 3
·Esters (酯) 5
----------------------------------------------------------------
demo光盘可以要到，找负责化学信息学的陈利先生，他很热心的，只可惜数据要用ISIS/b

ase才能打开看

Dock screening Database
分子結構資料庫：這些資料庫都可做2D、3D結構和文數字搜尋
acdfind.db：化合物商業來源資料庫
mddr3d.db︰藥用化合物資料庫
cmc3d.db：藥用化合物資料庫
nci3d.db：美國癌症研究院分子結構資料庫
nciaids.db：美國癌症研究院抗癌和抗愛滋病分子結構資料庫
csd2d.db：英國劍橋晶體結構資料庫
csd2d.db：英國劍橋晶體結構資料庫
isismx.db：學習用分子結構資料庫

不知道哪位能知道以上数据库的下载地址

6. 要格式转化，请看
babel 是一个很好的格式转化软件
http://www.ccl.net/cca/software/SOURCES/C/babel/index.shtml

The Dundee PRODRG Server 也不错
http://davapc1.bioch.dundee.ac.uk/prodrg/

7. 开始介绍分子对接软件
分子对接（docking）相关软件
QXP(Flo+)
http://uwmml.pharmacy.wisc.edu/Flo/floindx.html

据说是最牛的，牛软件是没有介绍的，它是Flo program中的一部分，可以下载，但有密码

，不知道有没有人能down下来
Glide
http://www.schrodinger.com/Products/glide.html

Glide is a fast and accurate docking program that addresses a number of proble

Glide is a fast and accurate docking program that addresses a number of proble

ms, ranging from fast database screening to highly accurate docking. Binding m

ode studies show a remarkable accuracy of 1.3 Å average RMS deviation re

lative to co-crystallized drug-sized structures in the Gold test set. In datab

ase screening, Glide performs consistently well over a wide range of receptor

types, with enrichment factors of 53 for Estrogen Receptor and 66 for Sugar-Bi

nding Protein.

评价不错的软件

dock
http://www.cmpharm.ucsf.edu/kuntz/

DOCK addresses the problem of "docking" molecules to each other. It explores w

ays in which two molecules, such as a drug and an enzyme or protein receptor,

might fit together. Compounds which dock to each other well, like pieces of a

three-dimensional jigsaw puzzle, have the potential to bind. So, why is it imp

ortant to able to identify small molecules which may bind to a target macromol

ecule? A compound which binds to a biological macromolecule may inhibit its fu

nction, and thus act as a drug.

DOCK generates many possible orientations (and more recently, conformations) o

f a putative ligand within a user-selected region of a receptor structure. The

Glide is a fast and accurate docking program that addresses a number of proble

ms, ranging from fast database screening to highly accurate docking. Binding m

ode studies show a remarkable accuracy of 1.3 Å average RMS deviation re

lative to co-crystallized drug-sized structures in the Gold test set. In datab

ase screening, Glide performs consistently well over a wide range of receptor

types, with enrichment factors of 53 for Estrogen Receptor and 66 for Sugar-Bi

nding Protein.

评价不错的软件

dock
http://www.cmpharm.ucsf.edu/kuntz/

DOCK addresses the problem of "docking" molecules to each other. It explores w

ays in which two molecules, such as a drug and an enzyme or protein receptor,

might fit together. Compounds which dock to each other well, like pieces of a

three-dimensional jigsaw puzzle, have the potential to bind. So, why is it imp

ortant to able to identify small molecules which may bind to a target macromol

ecule? A compound which binds to a biological macromolecule may inhibit its fu

nction, and thus act as a drug.

DOCK generates many possible orientations (and more recently, conformations) o

f a putative ligand within a user-selected region of a receptor structure. The

f a putative ligand within a user-selected region of a receptor structure. The

se orientations may be scored using several schemes designed to measure steric

and/or chemical complementarity of the receptor-ligand complex. These scores

may be used to evaluate likely orientations of a single ligand, or to rank mol

ecules from a database.
这可是分子对接软件的鼻祖，不过已经不是最好的了

Autodock
http://www.scripps.edu/pub/olson-web/doc/autodock/

AutoDock is a suite of automated docking tools. It is designed to predict how

small molecules, such as substrates or drug candidates, bind to a receptor of

known 3D structure.
It has applications in:

X-ray crystallography;
structure-based drug design;
lead optimization;
virtual screening (HTS)
combinatorial library design
protein-protein docking
chemical mechanism studies

AutoDock actually consists of three separate programs: AutoDock performs the d

ocking of the ligand to a set of grids describing the target protein; AutoGrid

pre-calculates these grids; and AutoTors sets up which bonds will treated as

rotatable in the ligand.

In addition to docking, the atomic affinity grids can be visualised. This can

help , for example, to guide organic synthetic chemists design better binders.

We have also developed a GUI called AutoDockTools, or ADT for short, which hel

ps to set up and analyze dockings. We have an automated packager to help you d

ownload all the parts of ADT.

GOLD
http://www.ccdc.cam.ac.uk/prods/gold/index.html

GOLD is a program for predicting how flexible molecules will bind to proteins

- a problem of enormous significance in the rational design of drugs and combi

natorial libraries.

GOLD Features:
a genetic algorithm methodology for protein-ligand docking
a genetic algorithm methodology for protein-ligand docking
full ligand and partial protein flexibility
energy functions partly based on conformation and non-bonded contact informati

on from the CSD.
Choice of scoring functions in version 2.0: GoldScore and ChemScore
A range of constraints

可免费试用

FlexX
FlexX  http://cartan.gmd.de/flexx/
FlexX  WWW  Interface  http://cartan.gmd.de/flexx//html/flexx-interface.shtml

FlexX is a computer program for predicting protein-ligand interactions. For a

protein with known three-dimensional structure and a small ligand molecule, Fl

exX predicts the geometry of the protein-ligand complex and estimates the bind

ing affinity.
The two main applications of FlexX are complex prediction and virtual screenin

g. Complex prediction is used, when you have a protein and a small molecule bi

nding to it but no structure of the protein-ligand complex. FlexX can be used

to create and rank a series of possible protein-ligand complexes. In virtual s

creening, you have a protein and a set of compounds and you are interested in

creening, you have a protein and a set of compounds and you are interested in

prioritizing the compounds for experimental testing.

FlexX Features
FlexX is fast. On our protein-ligand benchmark set with about 200 complexes, F

lexX takes about 70 seconds per complex prediction on average. The computing t

ime depends on the size of the active site, the size of the ligand, and the de

gree of ligand symmetry and lies in the range from a few seconds up to twenty

minutes.
The conformational flexibility of the ligand is considered. FlexX considers to

rsion angle flexibility as well as the conformational flexibility of ring syst

ems. In addition, enantiomerism can be handled as a degree of freedom which is

important in cases where the set of compounds are automatically created from

a 2D database. In the current version of FlexX, the protein is considered rigi

d.
The placement algorithm in FlexX is based on the interactions occurring betwee

n the molecules. This ensures that the search is limited to low-energy structu

res improving the quality of the results in a given amount of computing time.

FlexX is based on sophisticated physico-chemical models. The MIMUMBA torsion a

ngle database is used for the creation of conformers, an interaction geometry

database is used to exactly describe intermolecular interaction patterns. For

scoring, the Boehm function (with minor adaptions necessary for docking) is ap

scoring, the Boehm function (with minor adaptions necessary for docking) is ap

plied.

FlexX Future
FlexX is under ongoing development. We are working on several aspects of prote

in-ligand docking such as:
Time-efficient docking of combinatorial libraries
Using FlexX in parallel on a heterogeneous workstation cluster
Improving the quality of the described protein-ligand interactions and the sco

ring

Gramm
Gramm
http://reco3.ams.sunysb.edu/gramm/

GRAMM is a program for protein docking. To predict the structure of a complex,

it requires only the atomic coordinates of the two molecules (no information

about the binding sites is needed). The program performs an exhaustive 6-dimen

sional search through the relative translations and rotations of the molecules

. The molecular pairs may be: two proteins, a protein and a smaller compound,

two transmembrane helices, etc. GRAMM may be used for high-resolution molecule

s, for inaccurate structures (where only the gross structural features are kno

wn), in cases of large conformational changes, etc.
wn), in cases of large conformational changes, etc.

The Global RAnge Molecular Matching (GRAMM) methodology is an empirical approa

ch to smoothing the intermolecular energy function by changing the range of th

e atom-atom potentials. The technique locates the area of the global minimum o

f intermolecular energy for structures of different accuracy. The quality of t

he prediction depends on the accuracy of the structures. Thus, the docking of

high-resolution structures with small conformational changes yields an accurat

e prediction, while the docking of ultralow-resolution structures will give on

ly the gross features of the complex. More information about the GRAMM methodo

logy is on our laboratory research page.

Platforms
GRAMM is compiled on SGI R10000, SGI R4000, SGI R4400, SGI R8000, Sun SPARC, I

BM RS6000, DECAlpha, and PC (Windows and Linux). Windows version must work on

all 32-bit flavors of the MS Windows operating system. Linux version was compi

led on RedHat with glibc2.0.

注册后可免费下载

Hex
http://www.biochem.abdn.ac.uk/hex/

Hex is an interactive protein docking and molecular superposition program. Cur

rently, Hex understands protein and DNA structures in PDB format. Hex was writ

ten by Dave Ritchie at the University of Aberdeen.

The main thing which distinguishes Hex from other macromolecular (i.e. protein

and DNA) docking programs and molecular graphics packages is its use of spher

ical polar Fourier correlations to accelerate docking calculations. The graphi

cal nature of Hex came about largely because I wanted to visualise the results

of such docking calculations in a natural and seamless way, without having to

export unmanageably many (and usually quite big) coordinate files to one of t

he many existing molecular graphics packages. For this reason, the graphical c

apabilities in Hex are relatively primitive, although these days one can do qu

ite a lot with a few calls to OpenGL. Nonetheless, if your main interest is in

modelling macromolecular docking, Hex may have something new to offer!

Hardware Requirements
Hex will run on most Silicon Graphics and Sun workstations and Linux PC machin

es. The Linux PC versions are available for RedHat versions 7.0, 7.2, 7.3 and

8.0, but these should run under other Linux distributions.

注册后免费下载
注册后免费下载

还有许多大型的商业软件包里都有相关的软件，下面的目录比较完整

Protein-protein (peptide) docking
3D-Dock Suite [FTDock, RPScore and MultiDock] (BioMolecular Modeling, Cancer R

esearch UK)
Bielefeld Protein Docking [detects geometrical and chemical complementarities

between surfaces of proteins and estimates docking positions]
BiGGER  [protein-docking algorithm; intergrated in chemera, a molecular graphi

cs and modeling program for studying protein structures and interactions] (Bio

Tecnol, S.A.)
DOT [computation of the electrostatic potential energy between two proteins or

other charged molecules] (San Diego Supercomputer Center)
HEX [protein docking and molecular superposition program] (University of Aberd

een)
GRAMM - Global Range Molecular Matching [see also the database of Protein-Prot

ein Decoys for the validation of energy functions and refinement procedures] (

SUNY)

Protein-ligand docking
Affinity (Accelrys Inc.)
AutoDock (The Scripps Research Institute)
AutoDock (The Scripps Research Institute)
CombiBUILD [fragment-based docking; program created to aid the design of combi

natorial libraries] (Sandia National Labs)
DockVision (University of Alberta)
FRED - Fast Rigid Exhaustive Docking [multiconformer docking program] (OpenEye

)
FlexiDock (Tripos)
FlexX  (BioSolveIt GmbH)
GLIDE (Schrödinger GmbH)
GOLD (CDSD)
HINT!  - Hydropathic Interactions (Virginia Commonwealth University)
LIGPLOT [program for automatically plotting protein-ligand interactions] (Univ

ersity College of London)
SITUS 2.0 [program package for modeling of atomic resolution structures into l

ow-resolution density maps] (Scripps Research Institute)
VEGA [calculation of ligand-receptor interaction energy] (Milan University)

Protein-protein and protein ligand docking
DOCK (UCSF Molecular Design Institute)
ICM-Dock (MolSoft LLC)

8. 最后要看一下3D结构，可以用下面这些可视化软件
DS ViewerPro 5.0
DS ViewerPro 5.0
最好的结构分子观看软件
有了这个软件，你还用cn3D么？

Accelrys' Discovery Studio™ ViewerPro (DS ViewerPro) combines state of t

he art molecular graphics with a full range of Windows® desktop integratio

n tools. Using the DS ViewerPro you can visualize and share molecular informat

ion in a clear and consistent way, whether it's by exchanging chemical files o

r by including dynamic chemical structures in Word documents, spreadsheets, or

presentations

http://www.accelrys.com/dstudio/ds_viewer/

Analyzing Molecular Models
Next: Accessing DataAnalysis and measurement tools help you to evaluate geomet

ry and understand chemistry. Discovery Studio™ ViewerPro automatically c

alculates atomic and molecular properties for molecules sketched or read in. M

easurement monitors can be created, all of which are completely dynamic. As yo

ur modeling environment changes, your monitors are automatically recalculated.

The DS ViewerPro includes additional analysis features. For example, you coul

d investigate potential hydrogen bonds or compute a solvent-accessible surface

d investigate potential hydrogen bonds or compute a solvent-accessible surface

for a protein structure.

Computational Features:
Calculate:Number of atoms
Molecular formula
Molecular weight
Exact molecular weight
Molecular volume
Partial charge
Measure:Distances
Angles
Dihedrals Angles
Compute:
3D geometry from 2D
Surfaces
Chirality
Protein secondary structure
Hydrogen bonds
Bumps (contacts between nonbonded atoms)

Rasmol
Rasmol
*Evaluating and ranking drug targets, including protein-protein interaction in

terfaces, designing strategies for rational drug design
*Screening virtual libraries of millions of compounds using the revolutionary

Molsoft flexible docking and scoring procedure.
*Identifying interaction hot-spots, i.e. the candidate amino-acid positions in

volved in protein-protein interactions
*Predicting loop conformations in proteins
*Designing proteins with desired properties
*Docking flexible peptides to proteins
*Designing peptides blocking protein-protein interactions
*2D to 3D conversion, analysis and clustering of large compound libraries,
*Predicting compound properties, building QSAR models, 3D pharmacophore constr

uction and search.

windows下的软件
上官方网站就知道功能有多么强大,不过一般用户没法得到

免费的icm资源
ICM拿不到,但还有些免费的资源可以用一下

Abagyan Lab
http://abagyan.scripps.edu/lab/web/man/frames.htm
据说是ICM的老家,里面有Homology Modeling Server ,可以免费使用,感觉还不错
据说是ICM的老家,里面有Homology Modeling Server ,可以免费使用,感觉还不错

Molsoft
http://www.molsoft.com/
官方网站上有两个在线小工具,PDB Viewer 和2D to 3D Molecular Converter 可以使用

ICM Lite
ICM的简化版,免费推出,不过现在官方网站不提供下载了,但可以用google或天网搜一下,说

不定还能找到

2.HyperChem 7
软件名称：HyperChem 7
公司主页：http://www.hyper.com/
平台：Windows
简介：图形界面，有半经验方法（AM1，PM3等），UHF，RHF和CI和7.0版新增加的密度泛函

。可进行单点能，几何优化，分子轨道分析，预测可见-紫外光谱，蒙特卡罗和分子力学计

算。主页同时提供试用版下载。

功能：
1. 结构输入和对分子操作。
2. 显示分子。
3. 化学计算。用量子化学或经典势能曲面方法，进行单点、几何优化和过渡态寻找计算。

3. 化学计算。用量子化学或经典势能曲面方法，进行单点、几何优化和过渡态寻找计算。

可以进行的计算类型有：单点能，几何优化，计算振动频率得到简正模式，过渡态寻找，

分子动力学模拟，Langevin动力学模拟，Metropolis Monte Carlo模拟。支持的计算方法

有：从头计算，半经验方法，分子力学，混合计算。
4. 可以用来研究的分子特性有：同位素的相对稳定性；生成热；活化能；原子电荷；HOM

O-LUMO能量间隔；电离势；电子亲和力；偶极矩；电子能级；MP2电子相关能；CI激发态能

量；过渡态结构和能量；非键相互作用能；UV-VIS吸收谱；IR吸收谱；同位素对振动的影

响；对结构特性的碰撞影响；团簇的稳定性。
5. 支持用户定制的外部程序。
6. 其它模块：RAYTRACE模块，RMS Fit，SEQUENCE编辑器，晶体构造器；糖类构造器，构

像搜寻，QSAR特性，脚本编辑器。
7. 新的力场方法：Amber 2，Amber 3，用于糖类的Amber，Amber 94，Amber 96。
8. ESR谱。
9. 电极化率。
10. 二维和三维势能图。
11. 蛋白质设计。
12. 电场。
13. 梯度的图形显示。
14. 新增功能：密度泛函理论(DFT)计算；NMR模拟；数据库；Charmm蛋白质模拟；半经验

方法TNDO；磁场中分子计算；激发态几何优化；MP2相关结构优化；新的芳香环图；交互式

参数控制；增强的聚合物构造功能；新增基组。

3.Triton
3.Triton
我的理解它可以做酶分子的进化,不只到说的对不对
TRITON
http://ncbr.chemi.muni.cz/triton/

The program TRITON is graphical tool for modelling protein mutants and assessm

ent of their activities. Protein mutants are modelled based on the wild type s

tructure by homology modelling using the external program MODELLER. Chemical r

eactions taking place in the mutants active site are modelled using the semi-e

mpirical quantum mechanic program MOPAC. Semi-quantitative predictions of muta

nts activities can be achieved by evaluating the changes in energies of the sy

stem and partial atomic charges of the active site residues during the reactio

n. The program TRITON offers graphical tools for the preparation of the input

data files, for calculation and for the analysis of the generated output data.

Implementation ensures the overall integrity of consecutive steps of the mode

lling of mutants and calculation of reaction coordinates, but the program can

also be used simply for combinatorial generation of multiple mutants by homolo

gy modelling.
------------------------------------------------
以下转自biolover
biohity: 我鼓励作者Martin Prokop在新版中支持Mopac2002(原来他只想支持mopac2000)

，他做到了。但因为是alpha版，有很多bugs。正一个个向他指出呢。
alchemy: 应该是好东西，遗憾的是好东西都要在irix/linux上跑
alchemy: 应该是好东西，遗憾的是好东西都要在irix/linux上跑
Bac: 我的认识是，MOPAC是用半经验的量子化学方法计算突变部位的变化的。输入是氨基

酸残基的改变（结构的改变）。运行后得到应该是突变部位原子能量和电荷的变化。进而

算出活化能的变化。不知道理解得对不对，我想最好还是请Alchemy再给我们说说。
qianfeng:这个软件是否是这样运行的,先要输入酶的晶体结构，再输入你要突变的位点，

它用modeller做homology modelling，得到突变后的酶分子结构，再用MOPAC计算突变部位

原子能量和电荷的变化。进而算出活化能的变化。
还有几点没搞清楚
1 酶催化的活性部位以及催化过程好象是要自己输入的，不是软件分析的？这样的话必须

对这个酶的性质非常的清楚
2 最后得到的是活化能的变化，并不是表观的酶活？
3 它应该可以用于酶分子的rational design，但好象只能预测酶活，能否预测其它性质，

如酶的稳定性？

4.CAChe Group 软件介绍
有CAChe 5.0 和 BioMedCAChe
CAChe Group 软件介绍
http://www.cachesoftware.com/
CAChe 5.0 for Windows is a leading computer-aided chemistry modeling package f

or experimental chemists conducting research in life sciences, materials and c

hemicals, as well as for undergraduate and graduate educators. The new Version

5.0 includes a new semiempirical method with 4x greater accuracy than current

methods, challenging even experimental accuracy; modeling of molecules with u

methods, challenging even experimental accuracy; modeling of molecules with u

p to 20,000 atoms; the inclusion of all main group elements in one semiempiric

al method; and more.

BioMedCAChe is the new computer-aided chemistry software package designed spec

ifically for bio- and medicinal chemists. The package aids researchers in disc

overing structure-activity relationships, optimizing leads by maximizing activ

ity, and improving the prediction of bioavailability. The power of the package

enables researchers to predict properties of compounds that have never been m

ade or properties that have never been measured. This prediction enables resea

rchers to select those novel compounds for testing or synthesis that are most

likely to be successful, saving valuable lab time and resources. BioMedCAChe a

lso offers users the ability to devise ways to calculate molecular properties

and calibrate them to experimental results.

均可下载免费试用

5.ChemOffice
ChemOffice Ultra 2002
http://chemstore.cambridgesoft.com/
ChemOffice, the chemistry software suite that redefined the chemist's desktop,

transforms your PC into a chemical publishing, modeling, and database worksta

tion.
methods, challenging even experimental accuracy; modeling of molecules with u

p to 20,000 atoms; the inclusion of all main group elements in one semiempiric

al method; and more.

BioMedCAChe is the new computer-aided chemistry software package designed spec

ifically for bio- and medicinal chemists. The package aids researchers in disc

overing structure-activity relationships, optimizing leads by maximizing activ

ity, and improving the prediction of bioavailability. The power of the package

enables researchers to predict properties of compounds that have never been m

ade or properties that have never been measured. This prediction enables resea

rchers to select those novel compounds for testing or synthesis that are most

likely to be successful, saving valuable lab time and resources. BioMedCAChe a

lso offers users the ability to devise ways to calculate molecular properties

and calibrate them to experimental results.

均可下载免费试用

5.ChemOffice
ChemOffice Ultra 2002
http://chemstore.cambridgesoft.com/
ChemOffice, the chemistry software suite that redefined the chemist's desktop,

transforms your PC into a chemical publishing, modeling, and database worksta

tion.
tion.

This ultimate suite includes ChemDraw Ultra 7.0, Chem3D Ultra 7.0 and ChemFind

er Ultra 7.0. It adds E-Notebook Ultra 7.0, BioAssay Pro 7.0, MOPAC, Gaussian

& GAMESS interfaces, ChemSAR Server Excel, CLogP, Purchasing for Excel, CombiC

hem/Excel, and even more, as well as the full set of ChemInfo databases, inclu

ding ChemACX & ChemACX-SC, The Merck Index and ChemMSDX, to ChemOffice Pro.

Top Ten New & Popular Features
E-Notebook Ultra 7.0 - (W) Makes electronic laboratory journals a reality. Sto

res collections of pages created with ChemDraw, Microsoft Excel and Word.
BioAssay Pro 7.0 - (W) Integrate chemical and biological data. Query by struct

ure or text with ChemFinder and set up Excel templates for reporting and graph

ing.
The Merck Index 13th Electronic Edition - (W) An encyclopedia of chemicals, dr

ugs and biologicals with over 10,000 monographs on single substances or groups

of related compounds.
ChemACX & ChemACX-SC 7.0 - (W) 300 catalogs from leading chemical suppliers, i

ncluding Sigma-Aldrich, Fisher, Acros, Alfa Aesar, Lancaster and TCI America,

providing rapid ordering information for over 450,000 products. ChemACX-SC is

a collection of fully searchable catalogs from leading screening compound supp

liers.
Polymer Draw - (W, M) Represent and manipulate polymers in ChemDraw.
Polymer Draw - (W, M) Represent and manipulate polymers in ChemDraw.
BioArt - (W, M) A palette of customizable common biochemistry symbols includin

g membranes, cellular structures and more.
ChemDraw/Excel 7.0 - (W) Display and perform calculations on up to 1,400 chemi

cal structures at a time in Excel.
Name=Struct - (W, M) Generate a ChemDraw structure by typing in systematic che

mical names for most substances.
Purchasing for Excel - (W) Create shopping lists of compounds in Excel for aut

omated chemical purchasing.
ChemFinder/Word - (W) Use ChemFinder as an extension of MS Excel and Word for

Windows to create structure searchable spreadsheets and documents with an embe

dded ChemDraw application.

--
╔═══════════════════╗
║★★★★★友谊第一  比赛第二★★★★★║
╚═══════════════════╝

※ 来源:·哈工大紫丁香 http://bbs.hit.edu.cn·[FROM: 202.118.229.*]

Chemistry 版 (精华区)