Ripple's Blog

HOWTO:Low Mem requirment of Ubuntu

2006-04-13T18:34:00.000+08:00

Ubuntu/Debian-Sarge
Mini-RAM HOWTO

How to install an
Ubuntu-Desktop on low memory systems

(Pentium II and III Processor, 32-256 MB RAM)

by Ingo LANTSCHNER (ingo@binonabiso.com)

Thomas HINTERBERGER contributed a chapter,
which details how to install iceWM on Debian-Sarge; scroll to the end
of this document, if you want to see it.

Last updated: 26 Feb
2005

IMPORTANT:
This document has been written for Ubuntu Warty - so it is somehow
outdated. Well, we have done some additions for Hoary but probably
you'll find better guides in these days. Here are some (we have not
tested them!):

A
guide written by matic.ahacic [at] gmail.com.
You can find it here: http://ubuntuforums.org/showthread.php?t=42873

And here is one from Colin Applegate: http://linus.yhspatriot.net/cs/docs/ubuntu_howto/UbuntuLite

and an other one: http://www3.telus.net/public/lusseau/gnu-linux-ppcDocs/minimalUbuntuInstall.html

Want to do a journey
to
Congo with Bino na Biso?

Check out
Kotambola! (only German yet - sorry!)

But you can write us, we
will answer in English :-))

More images from Congo: ../work/Fotos/index.html

What this Howto is about

Ubuntu installs by default
with the Gnome Desktop and other memory-intensive applications. So if
you do not have a fast and powerful machine (Pentium 4 and 512 MB of
RAM) your system will be quite slow. The following document explains
the steps, how to install Ubuntu and a GUI on a low memory system, so
that you can use it as a Workstation for your daily work. It is aimed
to the average SOHO-user (Small Office/Home Office).

Screenshot (Click to
enlarge)

Conventions:

# command	Type command as root (you can do a sudo su- before, in order to get the root-prompt. Or you put sudo before the command.
$ command	Type command as user.
Red text	Parts which should be improved. Please let me know, if you have usefull suggestions.

1 The Basesystem

Get Ubuntu CD-ROM see http://www.ubuntulinux.org/

After inserting the CD-ROM into the
drive, boot and when asked to press Enter, type custom before. (Not linux custom!)

Hint for Hoary; In the Hoary installer there is no
'custom' install option. It appears they've replaced it with 'server'
which is a minimum install.

The result will be a minimal-system
with less than 300 MB on the HD and only a textprompt (no GUI).

2 Postinstall the GUI

I did it in the following order, but
probably the order is not important:

$ sudo su -

This puts you into a root-shell (#), so no more sudo
is necesary.

# vi
/etc/apt/sources.list

(If
you are not familiar with vi you can use nano or
any other texteditor instead.)

Enable the universe-repository by removing the Hashmarks (=# (2 times))

#
apt-get update

#
apt-get install icewm

#
apt-get install xserver-xfree86

#
apt-get install x-window-system-core

#
apt-get install xdm

#
apt-get install numlockx

#
apt-get install xterm

The result is a system with X and iceWM
as windowmanager. You log in as user and on the prompt:

$ startx

starts the GUI. (After the first
reboot, xdm autostarts and puts you directly into the GUI-mode)

So far this system needs 468 MB on your
harddrive.

From Frank Martelli [frank at
foodsavvy dot com]: Just a note when
dealing with the ppc version
of Ubuntu -- for some

reason, icewm (silently) requires /usr/lib/libtiff.so.3. A very
frustrating bug, as X starts lets you login and then (without any
errors)

returns to the login screen.

Solution:

ln -s /usr/lib/libtiff.so.4 /usr/lib/libtiff.so.3

3 Basic Applications

We consider
a Mailreader, Webbrowser, PDF-Reader and Officesuite as basic:

PDF-Reader

#
apt-get install acroread

This adds 26 MB to the harddrive
(compared to 3,2 MB of gPDF, but gPDF laks crucial features like search
and copy of text.)

Mozilla

We include Flashplayer and
Acrobat-Plugin for optimal compatibility with todays WWW. (The plugins
only need 2,5 MB)

#
apt-get install mozilla flashplayer-mozilla acroread-plugin

Fills up the harddrive to 555MB

Officesuite

At the date of writing this Howto you
have two options. Choose one of them (either-or, not
both!). This will probably change constantly, as OpenOffice envolves -
but the principle will stay the same (just the versions will change).

Either: Install OpenOffice 1.1.2 from apt-repository

#
apt-get install openoffice.org

This installs Openoffice 1.1.2

Or: Install OpenOffice 1.1.3 from OpenOffice.org:

I installed OpenOffice 1.1.3 from the
tar, provided by http://www.openoffice.org/.
After untaring the tree I cd into the tree and type:

#
./install

3.3 Final thoughts

Finally we have now 770 MB for the
bare system on the harddrive, so if we consider to have 128 MB minimum
for the swap, we need harddrives between 1 and 1,5 GB for such
Ubuntu-Desktops.

4 More Applications

Application	Installation	Size on HD
openssh-server (sshd)	apt-get install openssh-server	594 kB
gthumb	apt-get install gthumb	60,2 MB
Acrobat Reader	apt-get install acroread	26 MB
gpdf	apt-get install gpdf (Warning: gpdf has no search-feature and you can not copy&paste parts of text out of PDFs)	3,2 MB
OOo, German Language Pack	apt-get install openoffice.org-l10n-de	18,6 MB
OOo, German Help	apt-get install openoffice.org-help-de	23,5 MB
CUPS Daemon	apt-get install cupsys	23,6 MB
CUPS-Manager	apt-get install gnome-cups-manager	1,6 kB
Printerdrivers	apt-get install cupsomatic-ppd	22 MB (This is too much - any ideas how to reduce this?!)
The Gimp	apt-get install gimp	40,5 MB
cdrecord	apt-get install cdrecord	1,2 MB
mkisofs	apt-get install mkisofs	860 kB
nedit (Editor)	apt-get install nedit	3,3 MB
alicq (ICQ-Client)	apt-get install alicq	7,8 kB
gaim (ICQ-Client)	apt-get install gaim	18,4 MB
guitar (Dateikomprimierer)	apt-get install guitar	3,3 MB
zip und unzip	apt-get install zip apt-get install unzip	550 kB
rdesktop	apt-get install rdesktop	320 kB
tsclient (Graphisches Frontend für rdesktop und VNC)	apt-get install tsclient	940 kB
cdrtoaster Does not work yet!	apt-get install cdrtoaster	106 kB
Simple cdrx Does not work yet !	apt-get install simplecdrx	5,1 MB
gtoaster Does not work yet!	apt-get install gtoaster	6 MB
xine-ui (Multi Media Player)	apt-get install xine-ui	12 MB
w32codecs	apt-get install w32codecs	28,4 MB
MPlayer Does not work yet!	apt-get install mozilla-mplayer	18,3 MB
	apt-get install mplayer-fonts	7,3 MB
	apt-get install gstreamer0.8-plugins	11 MB
mc (Midnightcommander, Filemanager)	apt-get install mc	5 MB
emelfm (Filemanager)	apt-get install emelfm	582 kB

Todos

Backup

Find a working frontend for cdrecord.

Find a working Mplayer-package.

4 Future plans

If we go down to 64 or 32 MB of RAM we
won't install OpenOffice. I recommend Abiword and Gnumeric instead.
Also the Mozialla web broser could be a problem on such systems. Better
to go with Firefox and Thunderbird/Pine.

5 Comparison

If we compare Ubuntu's ability to adapt
to low-memory-systems with Fedora we will be surprised:

Installation-Typ	Fedora Core 3 [MB]	Ubuntu 4.10 [MB]
Minimal standard	571	298
Minimal trimmed (see http://www.simpaticus.com/linux)	430	??
w/ X, iceWM	??	450
w/ Mozilla	??	add. 58
w/ OpenOffice 1.1.3	??	add. 184

6 More help

Install and configure Ubuntu: http://ubuntuguide.org/

Installing Low-Memory-Systems: http://www.rule-project.org/

7 Feedback

Please send comments and Feedback to ingo@binonabiso.com.

8 Alternative Installation with Debian Sarge

by Thomas HINTERBERGER

After installing, I had the feeling,
that the Debian Icewm has some problems with the Ubuntu System (tar was
not working, the same with Acroread and OpenOffice - the paths looked
different). Not knowing Icewm very well, I just decided to start from a
Debian Netinstall CD-Image (even floppy-disks are possible - you
download it from www.debian.org ).
It is very small - including X with Icewm 298 MB. I think, that Icewm
works better with the original Debian - after installation all programs
was working immediately (but perhaps it is just the newer Version of
Icewm - 2.20 - you also get Firefox 1.0, Mozilla 1.7.5 and gimp 2.2)

you can follow the guide from Ingo - I
will note only the steps you have to add.

after installation you have no "sudo
su"- so just type su and the root-password to get root.

add the follwing lines to your
sources.list

deb ftp://ftp.debian.org/debian/ sarge main deb ftp://ftp.debian.org/debian-non-US/ sarge/non-US main deb-src ftp://ftp.debian.org/debian/ sarge main deb-src ftp://ftp.debian.org/debian-non-US/ sarge/non-US main

# netselect-apt finds the most quick of all
Debian-mirrors.

after # apt-get install
xterm you add

#
apt-get install sudo

#
apt-get install emelfm

when you are now the
first time in Icewm, verify with emelfm or with the shell, that there
is a file /etc/sudoers - if not, you have to reboot. - rootshell: #
shutdown -r now

#
echo "[username] ALL = NOPASSWD: /sbin/shutdown, /sbin/poweroff,

/sbin/halt, /sbin/reboot, /bin/cdrecord" >> /etc/sudoers

#
shutdown -r now

after rebooting you
should now be able to use all commands, when you press strg+alt+del. If
you don't do this, you have to write all the time # shutdown -r now to
a rootshell, to shut down the system proper.

instead of this, you can
edit /etc/sudoers, but only with the command

#
visudo

!!!! don't use another
editor !!!!- for visudo: strg o = saving, strg x = exit

you should have the
following lines:

root ALL=(ALL) ALL [username] ALL = NOPASSWD: /sbin/shutdown,
/sbin/poweroff, /sbin/halt, /sbin/reboot, /bin/cdrecord

The rest is the same like in Ingos description - the only thing - I
would use bluefish instead of nedit - it is 4 MB more, but you have a
wonderful HTML - Editor.

The configuration of X is not as good and comfortable, as with
Ubuntu - I did a little trick: run the Ubuntu live CD or Knoppix (if
you have only 64MB, you have to do a swap partition before - I did not
test it with Ubuntu, but Knoppix runs with 64MB) - make the screen and
fonts well looking- copy the /etc/X11/XF86Config-4 from the live CD to
a disk (or mail it), bring it to your /home directory in Icewm- save
your existing /etc/X11/XF86Config-4.

# cp XF86Config-4 /etc/X11/XF86Config-4

with my system, it was working perfect (both - Knoppix-config and
Ubuntu-config)

one more tip: emelfm is working as a file browser, like Nautilus or
Konqueror - it is not visible from the first moment, it is hard
to configure, but it does.

Back to www.binonabiso.com/en/

or www.binonabiso.com/de/

1 2/1/05

http://web.indstate.edu/thcme/mwking/protein-modifications.html

2006-02-27T10:45:00.000+08:00

Post-Translational Modification
Secreted and Membrane-Associated Proteins
Proteolytic Cleavage of Proteins
Glycoproteins
Mechanism of Sugar Linkage
Lysosomal Targeting of Enzymes
Clinical Significances of Glycoproteins
Defects in Glycoprotein Degradation
Acylation
Methylation
Phosphorylation
Sulfation
Prenylation
Vitamin C-Dependent Modifications
Vitamin K-Dependent Modifications
Selenoproteins Return to Medical Biochemistry Page

Secreted and Membrane-Associated Proteins
Proteins that are membrane bound or are destined for excretion are synthesized by ribosomes associated with the membranes of the endoplasmic reticulum (ER). The ER associated with ribosomes is termed rough ER (RER). This class of proteins all contain an N-terminus termed a signal sequence or signal peptide. The signal peptide is usually 13-36 predominantly hydrophobic residues. The signal peptide is recognized by a multi-protein complex termed the signal recognition particle (SRP). This signal peptide is removed following passage through the endoplasmic reticulum membrane. The removal of the signal peptide is catalyzed by signal peptidase. Proteins that contain a signal peptide are called preproteins to distinguish them from proproteins. However, some proteins that are destined for secretion are also further proteolyzed following secretion and, therefore contain pro sequences. This class of proteins is termed preproproteins.
Mechanism of synthesis of membrane bound or secreted proteins. Ribosomes engage the ER membrane through interaction of the signal recognition particle, SRP in the ribosome with the SRP receptor in the ER membrane. As the protein is synthesized the signal sequence is passed through the ER membrane into the lumen of the ER. After sufficient synthesis the signal peptide is removed by the action of signal peptidase. Synthesis will continue and if the protein is secreted it will end up completely in the lumen of the ER. If the protein is membrane associated a stop transfer motif in the protein will stop the transfer of the protein through the ER membrane. This will become the membrane spanning domain of the protein.back to the top

Proteolytic Cleavage
Most proteins undergo proteolytic cleavage following translation. The simplest form of this is the removal of the initiation methionine. Many proteins are synthesized as inactive precursors that are activated under proper physiological conditions by limited proteolysis. Pancreatic enzymes and enzymes involved in clotting are examples of the latter. Inactive precursor proteins that are activated by removal of polypeptides are termed, proproteins.
A complex example of post-translational processing of a preproprotein is the cleavage of prepro-opiomelanocortin (POMC) synthesized in the pituitary (see the Peptide Hormones page for discussion of POMC). This preproprotein undergoes complex cleavages, the pathway of which differs depending upon the cellular location of POMC synthesis.
Another is example of a preproprotein is insulin. Since insulin is secreted from the pancreas it has a prepeptide. Following cleavage of the 24 amino acid signal peptide the protein folds into proinsulin. Proinsulin is further cleaved yielding active insulin which is composed of two peptide chains linked togehter through disulfide bonds..
Still other proteins (of the enzyme class) are synthesized as inactive precursors called zymogens. Zymogens are activated by proteolytic cleavage such as is the situation for several proteins of the blood clotting cascade. back to the top

Glycoproteins
Membrane associated carbohydrate is exclusively in the form of oliogsaccharides covalently attached to proteins forming glycoproteins, and to a lesser extent covalently attached to lipid forming the glycolipids. Glycoproteins consist of proteins covalently linked to carbohydrate. The predominant sugars found in glycoproteins are glucose, galactose, mannose, fucose, GalNAc, GlcNAc and NANA. The distinction between proteoglycans and glycoproteins resides in the level and types of carbohydrate modification. The carbohydrate modifications found in glycoproteins are rarely complex: carbohydrates are linked to the protein component through either O-glycosidic or N-glycosidic bonds. The N-glycosidic linkage is through the amide group of asparagine. The O-glycosidic linkage is to the hydroxyl of serine, threonine or hydroxylysine. The linkage of carbohydrate to hydroxylysine is generally found only in the collagens. The linkage of carbohydrate to 5-hydroxylysine is either the single sugar galactose or the disaccharide glucosylgalactose. In ser- and thr-type O-linked glycoproteins, the carbohydrate directly attached to the protein is GalNAc. In N-linked glycoproteins, it is GlcNAc.
O-linkage to GalNAc
N-linkage to GlcNAc
The predominant carbohydrate attachment in glycoproteins of mammalian cells is via N-glycosidic linkage. The site of carbohydrate attachment to N-linked glycoproteins is found within a consensus sequence of amino acids, N-X-S(T), where X is any amino acid except proline. When an analysis of proteins in the public databases is carried out, it can be shown that approximately 65% of all the proteins contain at least one occurrence of the Asn-X-Ser/Thr consensus. N-linked glycoproteins all contain a common core of carbohydrate attached to the polypeptide. This core consists of three mannose residues and two GlcNAc. A variety of other sugars are attached to this core and comprise three major N-linked families:
1. High-mannose type contains all mannose outside the core in varying amounts.
2. Hybrid type contains various sugars and amino sugars.
3. Complex type is similar to the hybrid type, but in addition, contains sialic acids to varying degrees.
Structures of oligosaccharides of the 3 major classes of glycoprotein.Open squares: GlcNAc; open circles: mannose; open diamonds: galactose; filled squares: fucose; filled triangles: sialic acid the greek symbols a and b followed by numbers refers to the type of linkage.
Most proteins that are secreted, or bound to the plasma membrane, are modified by carbohydrate attachment. The part that is modified, in plasma membrane-bound proteins, is the extracellular portion of the protein that is modified. Intracellular proteins are less frequently modified by carbohydrate attachment. However, the attachment of carbohydrate to intracellular proteins confers unique functional activities on these proteins. Linkage of carbohydrate to cytosolic and/or nuclear proteins occurs via O-linkage and involves attachment of GlcNAc to serine or threonine residues. The linkage is catalyzed by the enzyme O-GlcNAc transferase, OGT. Several transcription factors and RNA polymerase II have been shown to be modified by O-GlcNAc linkage. back to the top

Mechanism of Carbohydrate Linkage to Protein
The protein component of all glycoproteins is synthesized from polyribosomes that are bound to the endoplasmic reticulum (ER). The processing of the sugar groups occurs cotranslationally in the lumen of the ER and continues in the Golgi apparatus for N-linked glycoproteins. Attachment of sugars in O-linked glycoproteins occurs post-translationally in the Golgi apparatus. Sugars used for glycoprotein synthesis (both N-linked and O-linked) are activated by coupling to nucleotides. Glucose and GlcNAc are coupled to UDP and mannose is coupled to GDP.
O-linked sugars: The synthesis of O-linked glycoproteins occurs via the stepwise addition of nucleotide-activated sugars directly onto the polypeptide. The nucleotide-activated sugars are coupled to either UDP, GDP (as with mannose) or CMP (for instance, NANA). The attachment of sugars is catalyzed by specific glycoprotein glycosyltransferases. Evidence indicates that each specific type of carbohydrate linkage in O-linked glycoproteins is the result of a different glycosyltransferase.
N-linked sugars: As indicated earlier, the three major classes of N-linked carbohydrate modifications are high-mannose, hybrid and complex. The major distinguishing feature of the complex class is the presence of sialic acid, whereas the hybrid class contains no sialic acid.
In contrast to the step-wise addition of sugar groups to the O-linked class of glycoproteins, N-linked glycoprotein synthesis requires a lipid intermediate: dolichol phosphate. Dolichols are polyprenols (C80-C100) containing 17 to 21 isoprene units, in which the terminal unit is saturated.
DolicholThe black bracket denotes the isoprene unit.The phosphate in dolichol phosphate is attached to the hydroxyl.
As indicated, the formation of the GlcNAc-b-Asn linkage in proteins occurs in the endoplasmic reticulum (ER) through cotranslational addition of a preassembled carbohydrate core structure that is delivered via the carbohydrate-dolichol lipid intermediate. The preassembled carbohydrate core structure comprises three terminal residues of glucose attached to a branched cluster of nine mannose residues that are in turn attached to two GlcNAc residues attached to dolicholpyrophosphate. The structure is abbreviated Glc3Man9GlcNAc2-P-P-dolichol. This structure is commonly referred to as the lipid-linked oligosaccharide (LLO), whereas the oligosaccharide structure itself is termed the en bloc oligosaccharide. In mammalian cells the importance of the terminal glucose residues is evident from the fact that transfer of Man9GlcNAc2-P-P-dolichol is some 25-times less efficient than the complete structure. In addition, structures that contain three terminal glucose residues, but not the complete Man9GlcNAc2 structure, are efficiently transferred to protein by oligosaccharyltransferase. Synthesis of the en bloc dolichol-P-P-oligosaccharide unit begins on the cytoplasmic face of the ER membrane and prior to transfer to the protein, the structure 揻lips?to the luminal side.
Pathway by which the synthesis and transfer of the lipid-linked oligosaccharide unit takes place at the membrane of the ER.
Immediately following transfer of the en bloc oligosaccharide unit to the protein, processing and alteration of the composition of the oligosaccharide ensues and continues as the protein passes through the ER then into and through the Golgi apparatus. Initially, the terminal glucose is removed through the action of glucosidase I (GI), a membrane bound enzyme recognizing a1,2-linked glucose. The remaining two glucose residues are then removed by glucosidase II (GII), a soluble enzyme recognizing a1,3-inked glucose. After removal of the glucose residues, the action of a-mannosidases removes several mannose residues as the protein progresses to the Golgi. The action of the various glucosidases and mannosidases leaves N-linked glycoproteins containing a common core of carbohydrate consisting of three mannose residues and two GlcNAc. Through the action of a wide range of glycosyltransferases and glycosidases a variety of other sugars are attached to this core as the protein progresses through the Golgi. These latter reactions generate the three major families of N-linked glycoproteins described above. back to the top

Lysosomal Targeting of Enzymes
Enzymes that are destined for the lysosomes (lysosomal enzymes) are directed there by a specific carbohydrate modification. During transit through the Golgi apparatus a residue of GlcNAc-1-phosphate (GlcNAc-1-P) is added to the carbon-6 hydroxyl group of one or more specific mannose residues that have been added to these enzymes. The GlcNAc is activated by coupling to UDP and is transferred by UDP-GlcNAc:lysosomal enzyme GlcNAc-1-phosphotransferase (GlcNAc-phosphotransferase), yielding a phosphodiester intermediate: GlcNAc-1-P-6-Man-protein. A second reaction (catalyzed by GlcNAc 1-phosphodiester-N-acetylglucosaminidase) removes the GlcNAc leaving mannose residues phosphorylated in the 6 position: Man-6-P-protein. A specific Man-6-P receptor (MPR) is present in the membranes of the Golgi apparatus. Binding of Man-6-P to this receptor targets proteins to the lysosomes.
Two distinct MPRs have been identified and both are members of the P-type lectin family. Both are type I integral membrane glycoproteins that contain an N-terminal extracellular domain, a single transmembrane domain and a C-terminal cytoplasmic domain. One receptor is large with a molecular weight of approximately 300kDa, the other receptor is smaller with a molecular weight of approximately 46kDa. Structural similarities between these two receptors indicates they are derived from a single ancestral gene with the larger receptor arising through multiple gene duplications. The extracellular portion of the larger receptor contains 15 repeating elements, each of which is highly similar to the extracellular domain of the smaller receptor. Both receptors exist as dimers embedded in the membrane.
The large receptor binds two moles of Man-6-P and the smaller binds one mole of Man-6-P per subunit, thus 4 and 2 moles of Man-6-P per dimer, respectively. The bovine and murine versions of the smaller receptors require divalent cations for ligand binding and thus the receptor has been termed the cation-dependent Man-6-P receptor (CD-MPR). However, the human counterpart may not require cations for its activity. The larger receptor does not require divalent cations for ligand binding and is therefore, commonly referred to as the cation-independent Man-6-P receptor (CI-MPR). However, the CI-MPR has been shown to bind the nonglycosylated polypeptide hormone, insulin-like growth factor II (IGF-II) and as such the larger MPR is more frequently identified as IGF-II/MPR. The IGF-II/MPR is available at the cell surface and its role in binding IGF-II is to target this hormone for degradation in the lysosomes. In addition to IGF-II, the IGF-II/MPR has been shown to bind a diverse array of Man-6-P-containing proteins as well as several nonglucosylated proteins. Although IGF-II/MPR and CD-MPR exhibit distinct activities, both receptors function to target newly synthesized lysosomal enzymes to the lysosomes. back to the top

Clinical Significances of Glycoproteins
Glycoproteins on cell surfaces are important for communication between cells, for maintaining cell structure and for self-recognition by the immune system. The alteration of cell-surface glycoproteins can, therefore, produce profound physiological effects, of which several are listed below.
1. The ABO blood group antigens are the carbohydrate moieties of glycolipids on the surface of cells as well as the carbohydrate portion of serum glycoproteins. When present on the surface of cells the ABO carbohydrates are linked to sphingolipid and are therefore of the glycosphingolipid class. When the ABO carbohydrates are associated with protein in the form of glycoproteins they are found in the serum and are referred to as the secreted forms. Some individuals produce the glycoprotein forms of the ABO antigens while others do not. This property distinguishes secretors from non-secretors, a property that has forensic importance such as in cases of rape. For more information of blood group antigens, including ABO visit the blood groups page at the SCARF site.
Structure of the ABO blood group carbohydrates,R represents the linkage to protein in the secreted forms, sphingolipid in the cell-surface bound form.open square = GlcNAc, open diamond = galactose, filled square = fucose, filled diamond = GalNAc, filled diamond = sialic acid (NANA)
2. The truncation of erythrocyte surface glycoproteins leads to cell clumping, as in congenital dyserythropoietic anemia type II. Also referred to as HEMPAS (hereditary erythroblastic multinuclearity with positive acidified-serum test).
3. Several viruses, bacteria and parasites have exploited the presence of cell-surface carbohydrates, principally associated with protein (glycoproteins), using them as portals of entry into the cell.
A. Human immunodeficiency virus (HIV), the causative agent of AIDS, gains entry into cells of the immune system by attaching to a class of cellular receptors known as the chemokine receptors, most notably CXCR4 and CCR5. For more information on chemokines and their receptors visit the C.O.P.E site.
B. Members of the poxvirus family of viruses gain entry into cells, most frequently migratory leukocytes, by attaching to chemokine receptors including CCR1, CCR5 and CXCR4 (Science [1999] vol. 286 pp. 1968-1971).
C. Dystroglycan (DG) is a component of the dystrophin-glycoprotein complex. It is a laminin receptor encoded by a single gene and cleaved by postranslational processing into two proteins, peripheral membrane a-DG and transmembrane b-DG. a-DG interacts with laminin-2 in the basal lamina and b-DG binds to dystrophin containing cytoskeletal proteins in muscle and peripheral nerves. DG is involved in agrin- and laminin-induced acetylcholine receptor clustering at neuromuscular junctions, morphogenesis, early development, and the pathogenesis of muscular dystrophies. Recent evidence (Science (1998) vol. 282 pp. 2076-2079 and 2079-2081) demonstrates that a-DG present on Schwann cell membranes is the receptor for Mycobacterium leprae and also serves as the receptor for the arenavirus class of pathogens. Arenaviruses cause hemorrhagic fever in humans. Lymphocytic choriomeningitis virus (LCMV), Lassa fever virus (LFV), Oliveros and Mobala (all members of the arenavirus family) all bind to a-DG. The specificity of this interaction was demonstrated by the resistance to LCMV infection of cells harboring a null mutation in DG.
D. Rhinoviruses utilize attachment to ICAM-1 (intercellular adhesion molecule-1) to gain entry into cells.
E. The pathogenic human parvovirus, B19, attaches to the erythrocyte-specific cell-surface globoside identified as erythrocyte P antigen to infect erythrocytes.
F. The malarial parasite Plasmodium vivax, binds to the erythrocyte chemokine receptor known as the Duffy blood group antigen (also known as the erythrocyte receptor for interleukin-8) to infect erythrocytes.
G. The MN blood group system is a well-characterized set of erythrocyte surface antigens that represent the variable carbohydrate modifications of the trans-membrane glycoprotein, glycophorin. Glycophorin is the cellular receptor for influenza virus as well as the receptor for erythrocyte invasion by the malarial parasite Plasmodium falciparum.
H. Helicobacter pylori is the bacterium responsible for chronic active gastritis and gastric and duodenal ulcers; it is also the causative agent for one of the most common forms of cancer in humans, adenocarcinoma. This bacterium attaches to the Lewis blood group antigen on the surfaces of gastric mucous cells.
I. Rabies virus binds to cells through interactions with neural cell adhesion molecule (N-CAM).
J. The receptor for fibroblast growth factor (FGF) has been reported to be the portal of entry for human herpes virus Type I. Recent new evidence indicates that the portal of entry for human herpes simplex Type I viruses is 3-O-sulfated heparan sulfate (Cell 99:13-22, 1999).
K. Human herpesvirus 6 (HHV-6) infection occurs in virtually all persons within the first 2 years of life and persists the entire lifetime. In immunocompromised patients HHV-6 causes opportunistic infections and is the causative agent of exanthema subitum. HHV-6 has been linked to multiple sclerosis and to the progression of AIDS. The cellular receptor for HHV-6 is the cell-surface type-I glycoprotein, CD46 (Cell 99:817-827, 1999).
4. Some glycoproteins are tethered to the membrane by a lipid linkage: the protein is attached to the carbohydrate through phosphatidylethanolamine (PE) linkage, and the carbohydrate is in turn attached to the membrane via linkage to phosphatidylinositol (PI), which anchors the structure within the membrane. The linkage is called a glycosylphosphotidylinositol (GPI) anchor, and proteins that are anchored in this way are termed glypiated proteins. The disease, paroxysmal nocturnal hemoglobinuria, results from the loss of the erythrocyte surface glycoprotein, decay-accelerating factor, (DAF). DAF prevents erythrocyte lysis by complement. When this factor is lost from the erythrocyte surface, abnormal hemolysis occurs, with the end result of hemoglobin accumulation in the urine.
The GPI linkage of the T-cell marker Thy-1Line represents the outer surface of the membraneSquiggles represent the lipid portion of the GPI linkage embedded in the membraneOpen circles = mannose, filled diamonds = GalNAc, filled squares = fucoseFilled pentagons = ethanolamine, solid circles with P = phosphates
Other important GPI linked proteins are the enzymes acetylcholinesterase, intestinal and placental alkaline phosphatase and 5'-nucleotidase, the cell adhesion molecule N-CAM (neural cell adhesion molecule) and the T-cell markers Thy-1 and LFA-3 (lymphocyte function associated antigen-3).
5. The proper degradation of glycoproteins has medical relevance. Degradation occurs within lysosomes and requires specific lysosomal hydrolases, termed glycosidases. Exoglycosidases remove sugars sequentially from the non-reducing end and exhibit restricted substrate specificities. In contrast, endoglycosidases cleave carbohydrate linkages from within and exhibit broader substrate specificities. Several inherited disorders involving the abnormal storage of glycoprotein degradation products have been identified in humans. These disorders result from defects in the genes encoding specific glycosidases, leading to incomplete degradation and subsequent over-accumulation of partially degraded glycoproteins. As a general class, such disorders are known as lysosomal storage diseases and include the diseases known as mucolipidoses that result from incomplete degradation of the carbohydrate portions of glycolipids.
6. Defects in the proper targeting of glycoproteins to the lysosomes can also lead to clinical complications. Deficiencies in the enzyme responsible for the transfer of GlcNAc-1-P to Man residues (GlcNAc phosphotransferase) in lysosomal enzymes leads to the formation of dense inclusion bodies formation in the fibroblasts. Two disorders related to deficiencies in the targeting of lysosomal enzymes are termed I-cell disease (mucolipidosis II) and pseudo-Hurler polydystrophy (mucolipidosis III, also called mucolipidosis-HI). I-cell disease is characterized by severe psychomotor retardation, skeletal abnormalities, coarse facial features, painful restricted joint movement, and early mortality. Pseudo-Hurler polydystrophy is less severe; it progresses more slowly, and afflicted individuals live to adulthood.
Enzyme Defects in Degradation ofAsn-GlcNAc Type Glycoproteins
Disease
Enzyme Deficiency
Symptoms/Comments
aspartylglycosaminuria
aspartylglycosaminidase
progressive mental retardation, delayed speech and motor development, coarse facial features
b-Mannosidosis
b-Mannosidase
primarily neurological defects, speech impairment
a-Mannosidosis
a-Mannosidase
mental retardation, dystosis multiplex, hepatosplenomegaly, hearing loss, delayed speech
GM1 Gangliosidosis
b-Galactosidase
also identified as a glycosphingolipid storage disease
GM2 Gangliosidosis(Sandhoff-Jatzkewitz disease)
b-N-acetylhexosaminidases A and B
also identified as a glycosphingolipid storage disease
Sialidosis(also identified as Mucolipidosis I)
Neuraminidase(sialidase)
myoclonus, congenital ascites, hepatosplenomegaly, coarse facial features, delayed mental and motor development
Fucosidosis
a-Fucosidase
progressive motor and mental deterioration, growth retardation, coarse facial features, recurrent sinus and pulmonary infectionsback to the top
OMIM links for additional Defects in Glycoprotein Degradation

Acylation
Many proteins are modified at their N-termini following synthesis. In most cases the initiator methionine is hydrolyzed and an acetyl group is added to the new N-terminal amino acid. Acetyl-CoA is the acetyl donor for these reactions. Some proteins have the 14 carbon myristoyl group added to their N-termini. The donor for this modification is myristoyl-CoA. This latter modification allows association of the modified protein with membranes. The catalytic subunit of cyclicAMP-dependent protein kinase (PKA) is myristoylated. back to the top

Methylation
Post-translational methylation occurs at lysine residues in some proteins such as calmodulin and cytochrome c. The activated methyl donor is S-adenosylmethionine. back to the top

Phosphorylation
Post-translational phosphorylation is one of the most common protein modifications that occurs in animal cells. The vast majority of phosphorylations occur as a mechanism to regulate the biological activity of a protein and as such are transient. In other words a phosphate (or more than one in many cases) is added and later removed.
Physiologically relevant examples are the phosphorylations that occur in glycogen synthase and glycogen phosphorylase in hepatocytes in response to glucagon release from the pancreas. Phosphorylation of synthase inhibits its activity, whereas, the activity of phosphorylase is increased. These two events lead to increased hepatic glucose delivery to the blood.
The enzymes that phosphorylate proteins are termed kinases and those that remove phosphates are termed phosphatases. Protein kinases catalyze reactions of the following type:
ATP + protein <----> phosphoprotein + ADP
In animal cells serine, threonine and tyrosine are the amino acids subject to phosphorylation. The largest group of kinases are those that phsophorylate either serines or threonines and as such are termed serine/threonine kinases. The ratio of phosphorylation of the three different amino acids is approximately 1000/100/1 for serine/threonine/tyrosine.
Although the level of tyrosine phosphorylation is minor, the importance of phosphorylation of this amino acid is profound. As an example, the activity of numerous growth factor receptors is controlled by tyrosine phosphorylation. back to the top

Sulfation
Sulfate modification of proteins occurs at tyrosine residues such as in fibrinogen and in some secreted proteins (eg gastrin). The universal sulfate donor is 3'-phosphoadenosyl-5'-phosphosulphate (PAPS).
Since sulfate is added permanently it is necessary for the biological activity and not used as a regulatory modification like that of tyrosine phosphorylation. back to the top

Prenylation
Prenylation refers to the addition of the 15 carbon farnesyl group or the 20 carbon geranylgeranyl group to acceptor proteins, both of which are isoprenoid compounds derived from the cholesterol biosynthetic pathway. The isoprenoid groups are attached to cysteine residues at the carboxy terminus of proteins in a thioether linkage (C-S-C). A common consensus sequence at the C-terminus of prenylated proteins has been identified and is composed of CAAX, where C is cysteine, A is any aliphatic amino acid (except alanine) and X is the C-terminal amino acid. In order for the prenylation reaction to occur the three C-terminal amino acids (AAX) are first removed and the cysteine activated by methylation in a reaction utilizing S-adenosylmethionine as the methyl donor.
Important examples of prenylated proteins include the oncogenic GTP-binding and hydrolyzing protein Ras and the g-subunit of the visual protein transducin, both of which are farnesylated. Numerous GTP-binding and hydrolyzing proteins (termed G-proteins) of signal transduction cascades have g-subunits modified by geranylgeranylation. back to the top

Vitamin C-Dependent Modifications
Modifications of proteins that depend upon vitamin C as a cofactor include proline and lysine hydroxylations and carboxy terminal amidation. The hydroxylating enzymes are identified as prolyl hydroxylase and lysyl hydroxylase. The donor of the amide for C-terminal amidation is glycine.
The most important hydroxylated proteins are the collagens. Several peptide hormones such as oxytocin and vasopressin have C-terminal amidation. back to the top

Vitamin K-Dependent Modifications
Vitamin K is a cofactor in the carboxylation of glutamic acid residues. The result of this type of reaction is the formation of a g-carboxyglutamate (gamma-carboxyglutamate), referred to as a gla residue.

Structure of a gla residue
The formation of gla residues within several proteins of the blood clotting cascade is critical for their normal function. The presence of gla residues allows the protein to chelate calcium ions and thereby render an altered conformation and biological activity to the protein. The coumarin-based anticoagulants, warfarin and dicumarol function by inhibiting the carboxylation reaction. back to the top

Selenoproteins
Selenium is a trace element and is found as a component of several prokaryotic and eukaryotic enzymes that are involved in redox reactions. The selenium in these selenoproteins is incorporated as a unique amino acid, selenocysteine, during translation. A particularly important eukaryotic selenoenzyme is glutathione peroxidase. This enzyme is required during the oxidation of glutathione by hydrogen peroxide (H2O2) and organic hydroperoxides.

Structure of the selenocysteine residue
Incorporation of selenocysteine by the translational machinery occurs via an interesting and unique mechanism. The tRNA for selenocysteine is charged with serine and then enzymatically selenylated to produce the selenocysteinyl-tRNA. The anticodon of selenocysteinyl-tRNA interacts with a stop codon in the mRNA (UGA) instead of a serine codon. The selenocysteinyl-tRNA has a unique structure that is not recognized by the termination machinery and is brought into the ribosome by a dedicated specific elongation factor. An element in the 3' non-translated region (UTR) of selenoprotein mRNAs determines whether UGA is read as a stop codon or as a selenocysteine codon. back to the top
Return to Medical Biochemistry Page
Michael W. King, Ph.D / IU School of Medicine / miking at iupui.edu
Last modified: Monday, 22-Aug-2005 08:08:08 EST

post-translational modifications

2006-02-27T10:44:00.000+08:00

Modifications
General Approach
Most protein samples exhibit some degree of modification.
There are the "natural" post translational modifications, such as phosphorylation and glycosylation. There are the accidental modifications which are artefacts of sample handling, such as oxidation. Finally, there are the modifications deliberately introduced during sample work-up, such as cysteine derivatisation. In most cases, it is only the deliberate modifications which are known about for certain at the time of doing a search.
It might be assumed that the search software could allow for those modifications which are described in sequence entry annotations. However, writing code to parse these sequence annotations would be a major task. Indeed, many post-translational modifications are not specified in a way which can be readily translated into specific mass differences. For example, noting that a residue is an actual or potential glycosylation site is not much help. Even a simple modification, such as phosphorylation, is rarely quantitative, so that it would be necessary to include mass values for all permutations of occupied and unoccupied sites.
And, of course, protein sequences derived translated from nucleotide sequences contain no information on post translational modifications.
The solution adopted here is to allow modifications to be specified in two different ways: fixed modifications and variable modifications.
Fixed modifications are applied universally, to every instance of the specified residue or terminus. There is no computational overhead associated with a fixed modification, it is simply equivalent to using a different mass for the modified residue or terminus. For example, selecting Carboxymethyl (C) means that all calculations will use 161 Da as the mass of cysteine.
Variable modifications are those which may or may not be present. Mascot tests all possible arrangements of variable modifications to find the best match. For example, if Oxidation (M) is selected, and a peptide contains 3 methionines, Mascot will test for a match with the experimental data for that peptide containing 0, 1, 2, or 3 oxidised methionine residues. This greatly increases the complexity of a search, resulting in longer search times and reduced specificity, so variable modifications should be used sparingly.
Unimod
The list of modifications used by Mascot is taken directly from the Unimod database. For further details of individual modifications, please refer to Unimod. Note that Unimod is an community supported resource. If you want to add a new modification to Unimod, you can do so, and you then become the curator of the new record. The Mascot modifications list is updated from Unimod each weekend.
Only selected modifications are displayed by default in the Mascot search form. If you want to see the complete list, you must go to the search form defaults page and tick the checkbox for 'Show all mods.'.
Other lists of modifications
DeltaMass is a comprehensive list of modifications, sorted by mass.
RESID database contains detailed descriptions of many post-translational modifications.

0x80248011

2006-02-02T14:09:00.000+08:00

When searching for available updates on the Windows Update site, you receive the 0x80248011 error. Problem descriptionThis error can occur due to an issue with the Internet cache or corruption in the Data store. Applicable operating systems
Windows 2000
Windows XP
Windows Server 2003
Resolutions
1. Delete the datastore and allow it to rebuild itself. First stop the Automatic Update Service
Click Start.
Choose Run.
In the Run box, type services.msc.
Click OK.
Right-click the Automatic Updates Service.
Click Stop. After Stopping the Service please rename the folder c:\Windows\SoftwareDistribution
Open Windows Explorer
Navigate to the Windows folder
Click on the + next to the Windows folder
Navigate to SoftwareDistribution folder
Right Click on the SoftwareDistribution folder
Select rename from the Menu
Rename the folder to SoftwareDistribution.old and click Enter Now restart the Automatic Update Service
Click Start.
Choose Run.
In the Run box, type services.msc.
Click OK.
Right-click the Automatic Updates Service.
Click Start. Starting the service will take a moment.Retry Windows Update againDid this resolve your problem?

2. Clear the Internet Explorer cache
Start Internet Explorer.
On the Tools menu, click Internet Options.
Click the General tab.
In the Temporary Internet files section, click Delete Cookies.
Click OK in the Delete Cookies box.
In the Temporary Internet Files section, click Delete Files.
Select the Delete all offline content check box in the Delete Files dialog box.
Click OK.
In the History section, click the Clear History button.
Click Yes.
Click OK.Did this resolve your problem?
For more information about When searching for available updates on the Windows

How Spyware Works

2006-01-26T14:10:00.000+08:00

How Spyware Works
In this paper, we'll briefly outline two systems that could be classified as spyware to demonstrate different methods for collecting information from users without their knowledge.2
In both cases, these systems perform some kind of surreptitious user tracking and then format some part of that data for reporting back to system's operator. It should be noted that there are significantly more egregious cases of spyware in use; we choose these two systems because they represent a Windows-based system that collects and reports information and a Web-based system to do the same. Other cases that we have analyzed include Spector Professoinal [5], TheCounter.com [2], Coremetrics [7], DoubleClick [8,9], and Netscape [6].
PCFriendly
PCFriendly is an application that shipped on numerous DVD titles between 1996 and 2000. In addition to its stated objective (providing a software-based DVD player for Windows machine), the system collected information about the user and the user's DVD collection, occasionally reporting such things back to InterActual Technologies, the maker of PCFriendly.
PCFriendly is a Windows-based application that starts when a DVD is inserted into the system's DVD player. The first time that the application starts, the user is asked for information like name, address, email address, and age. A unique identifier is assigned to the user, and the application appears to track changes over time, for example, additional DVD titles put into the system.
As of Interhack's last look at the system (in May 2002), PCFriendly was being replaced by a new system, known as InterActual Player, written to address privacy concerns, among other things.
Defenses that we identified at the time essentially meant breaking PCFriendly functionality:
Do not watch DVD titles on a computer. Backchannels are easy to implement on systems with Internet connectivity.
Upgrade to the latest InterActual Player. Privacy problems with Web-based content provided by the DVD producers (as opposed to InterActual Technologies) will not be addressed in this case, making this solution incomplete.
Block all access to pcfriendly.com and interactual.com domains.
Note that in our second option--using the InterActual Player--only meant using newer software designed to give the user a greater number of options for protecting privacy--default behavior was still invasive.
Detailed analysis and discussion is available in the Interhack Research Technical Report, “PCFriendly Enables DVD Backchannels”. [4]
Pharmatrak
Pharmatrak was a company that provided a Web site tracking and reporting service to pharmaceutical companies. Its system works much like the Coremetrics system Interhack analyzed in 2000 [7], with two critical differences. Interhack provided forensic analysis to plaintiffs' counsel in the Pharmatrak privacy litigation. Facts regarding the operation of Pharmatrak's service are identified in court documents. [10]
The first difference was that Pharmatrak did not have JavaScript code that was designed to pull users' responses to form data out of the form and to put them into a request for a Web bug. Pharmatrak's entire collection mechanism was predicated on collecting HTTP Referer [sic], though it did go to some significant lengths to get the data--including the use of JavaScript (and even a Java applet in the earliest instance of the software).
The Pharmatrak system was designed to collect information about users of pharmaceutical companies Web sites. The users would be pseudonymously tagged, and their activity observed and reported back to the pharmaceutical company. In addition, very high-level information (such as total traffic) would be reported to other pharmaceutical companies that were Pharmatrak clients, allowing each Pharmatrak client to see not only detailed information about its site activity, but high-level information about its competitors' sites. All of this happened with the knowledge and consent of the pharmaceutical companies that hired Pharmatrak to perform the reporting service and implanted the Pharmatrak-supplied code on their sites.
This leads us to our second difference: Pharmatrak was not authorized by its clients to collect personally-identifiable information, and by all appearances, Pharmatrak did not have specific intent to collect such information. (Forensic investigation and analysis showed that they did have detailed personal information on several hundred users.)
Detailed information on the mechanisms for client and server interaction on the Pharmatrak system can be found in court documents. Interaction between Web browsers and clients and how these impact user privacy is described in detail in Developing Trust: Online Privacy and Security. [3]

cite from http://web.interhack.com/publications/spyware_how.php

When the only tool you have is a hammer

2006-01-25T10:47:00.000+08:00

Every Problem you encounter tends to resemble a nail

-source unknown

Solaris Enterprise System Downloads

2006-01-17T21:02:00.000+08:00

Solaris Enterprise System Downloads

Get the Solaris Enterprise System

Download a complete enterprise-class solution or parts for multiple platforms—Solaris 10 OS, Java Enterprise System, development tools, N1 software—no cost, no kidding.

and it is free

GAME ON! Podcast

2006-01-17T21:00:00.000+08:00

GAME ON! Podcast
The business and culture of the video game industry
Discussions with industry luminaries on topics from game development and consoles, to the impact of video games on our culture and the future of electronic gaming.

Gamers and developers - listen up!
Hear first-hand, video game creators' experiences with retro games, video game music, hot picks - and how this form of entertainment has parlayed the consumer passion for video games into a multi-billion dollar industry.

Tune in, Game On, Have Fun!

Sparkle — Microsoft finally takes UI design seriously

2006-01-13T23:42:00.000+08:00

If you are into UI design and haven’t yet seen the video that Scoble posted on Microsoft’s Channel 9, prepare to crap your pants and go watch it.
At their Professional Developers Conference in Los Angeles today, Microsoft debuted Sparkle Interactive Designer as a part of their new Expression Suite. It’s a move that shows the Redmond crew seems to finally be taking UI design seriously.

Embeddable Java Based Reporting

2006-01-13T13:12:00.000+08:00

Learn how Style Report delivers a versatile Java based reporting environment that can be embedded inside desktop Java applications, deployed as a Java web application or used as central a reporting portal.

One click to download Style Report summary(2 pages PDF 140k)

More Resources

See Style Report's report delivery over web.

Go to live production reports.

Learn Style Report's zero client, self service, web reporting environment.

Go to the guided tour.

machine learning methods

2006-01-03T16:14:00.000+08:00

Software

Gaussian Processes

GP Demo. Demonstration Software for Gaussian Processes by David MacKay (in OCTAVE).

LS-SVMlab. Matlab/C toolbox for least squares support vector machines.

MAP-1. Package for MAP estimation by Carl Rasmussen.

MC-1. Package for MAP estimation by Carl Rasmussen.

Flexible Bayesian Modelling. Package by Radford Neal. It includes programs for Neural Networks, Gaussian Processes, and Mixture Models.

Netlab. Matlab toolbox including Gaussian Process Regression, Mixture models and Neural Networks.

Sparse Gaussian Processes. Matlab Toolbox for Sparse Inference using Gaussian Processes.

Tpros and Cpros. Package by Mark Gibbs.

Mathematical Programming

CPLEX. Barrier/QP Solver.

LOQO. Linear and Quadratic Optimization Package by Robert Vanderbei.

MINOS. Linear and Quadratic Solver.

Support Vectors

Nearest Point Algorithm. by Sathiya Keerthi (in FORTRAN).

SVM Java Applet. by Chris Burges et al.

BSVM. A decomposition method for bound-constrained SVM formulations.

QP SVM Classification and Regression. Fortran Implementation.

Chunking Code. by C. Saunders, M. O. Stitson, J. Weston, L. Bottou, B. Schölkopf, and A. Smola at Royal Holloway, AT&T, and GMD FIRST (Documentation).

cSVM. SVM for classification tasks with model selection.

DTREG. by Phillip H. Sherrod.

Interior Point Optimizer for SVM Pattern Recognition. by Alex Smola.

Equbits Foresight. Commerical SVM based Classification and Regression Application Designed for Drug Discovery.

Gini-SVM. A multi-class Probabilistic regression software for large data sets.

GiniSVM. Multi-class SVM Probability regression package.

Gist. Gist contains software tools for support vector machine classification and for kernel principal components analysis. The SVM portion of Gist is available via an interactive web server.

Parallel GPDT. Parallel and serial training of SVM.

HeroSvm1.0. A high performance DLL for training SVM on a very large training set efficiently.

SVM java implementation. This implementation is simple and easy to modify.

LEARNSC. SVM, NN and FL MATLAB based user-friendly routines.

LIBSVM. An SVM library with a graphic interface.

looms. a leave-one-out model selection software based on BSVM.

LS-SVMlab. Matlab/C Toolbox for Least Squares Support Vector Machines.

M-SVM. Multi-class support vector machine for very large problems.

M-SVM. Multi-class support vector machine for very large problems.

mySVM. SVM implementation for pattern recognition and regression.

mySVM and SVMlight for Windows. SVM implementation for Windows, uses Microsoft Visual C++ 6.0.

mySVM/db. SVM implementation to be run inside a database.

Sequential Minimal Optimization. by Xianping Ge.

SMOBR. SMOBR is an implementation of the original Sequential Minimal Optimisation proposed by Platt written in C++.

SVMdark. A Windows implementation of a support vector machine.

SvmFu. by Ryan Rifkin.

SVMLight. by Thorsten Joachims.

SVM/LOO. Matlab code for SVM incremental learning and decremental unlearning (LOO validation).

SVM/optima. SVM QP routines in Fortran for classification/regression.

SVMseq. An implementation of grad. desc. for SV learning, supports sample selection, string kernels and quasi-linear training. Implemented in Haskell, source + binaries available.

SVMTorch. Support Vector Machine for Large-Scale Regression and Classification Problems.

Matlab SVM Toolbox. by Steve Gunn.

Matlab SVM Toolbox. Matlab implementation in the style of SVMlight, can train 1-norm and 2-norm SVMs.

OSU SVM Classifier Matlab Toolbox. A matlab toolbox with a C++ mex core to fast implement the SVM classifiers.

SimpleSVM Toolbox. Fully Matlab toolbox for SVM, based on SimpleSVM algorithm. Includes 1class, invariance treatment.

SVM Toolbox. Object Oriented MATLAB Support Vector Machine Toolbox, including C++ MEX implementation of the sequential minimal optimisation algorithm.

WinSVM. SVM program for running under Windows.It uses SMO algorithm, so it is very fast and easy to use.

WinSVM. SVM for windows,easy to use.

winSVM. A Windows implementation of a support vector machine.

Other Algorithms

AdaBoost-Reg. A regularized version of the AdaBoost algorithm (in MATLAB).

Generalized Discriminant Analysis. Zip file, for Matlab 5.

Kernel Billiard. by Pal Rujan (in C).

Kernel ICA. A kernel-based approach for independent component analysis.

JINFIL java Instance Filtering. Instance Filtering is a preprocessing step for supervised learning systems for entity recognition in texts. The goal of Instance Filtering is to reduce both the skewed class distribution and the data set size by eliminating negative instances, while preserving positive ones as much as possible. This process is performed on both the training and test set, with the effect of reducing the learning and classification time, while maintaining or improving the prediction accuracy. The tool demonstrate excellent performances when applied to SVM classifiers.

Kernel-Machine Library. A GPL'ed C++ library to develop (new) kernel machine tools and algorithms in an efficient way.

myKLR. Kernel Logistic Regression.

RBF Networks. Fast RBF Networks with adaptive centers.

Kernel PCA. RBF Toy Example by Bernhard Schölkopf (in MATLAB).

R-KDDA. A regularized kernel discriminant analysis method (in matlab).

Spider. A library in MATLAB for classification, regression, clustering, .... for SVMs it uses LIBSVM and SVMLight.

Torch. a new machine learning library in C++/GPL including MLP, RBF, SVM, GMM, HMM, KNN, Parzen...

Java Development on Ubuntu

2005-12-30T14:59:00.000+08:00

The following are from David Coldrick's Blog and I found it is usefull

Netbeans-5_0-beta-bin-linux.bin, make sure JAVA_HOME is defined, and run the .bin. For example:

export JAVA_HOME=/usr/lib/j2sdk1.5-sun/
./netbeans-5_0-beta-bin-linux.bin

Note: If the installer can't find the jdk - because you haven't set JAVA_HOME - you might get the message: "The wizard cannot continue because of the following error: could not load wizard specified in /wizard.inf (104)". Hmmm. Could be deemed rather obscure.

All going well, not only have you installed a fantastic IDE, but you even have an eye-catching little double-clickable launcher on your desktop. Sweet.

Java Application Server: If the only J2EE you're doing is servlets and JSPs, you can use the Tomcat server builtin to NetBeans. However, if you want to get into EJBs and Web Services, install this: it's very well integrated into NetBeans, it's free, and the next version, which will be supported by a future version of NetBeans, is open source.

Download, choosing "Sun Java System Application Server Platform Edition 8.1 2005Q2 UR2". The file will be something like sjsas_pe-8_1_02_2005Q2-linux.bin.

Make sure it's executable (chmod +x sjsas_pe-8_1_02_2005Q2-linux.bin if necessary), and run it. It will probably not be able to find your copy of java, so you can do either:

sudo ./sjsas_pe-8_1_02_2005Q2-linux.bin -javahome /usr/lib/j2sdk1.5-sun

or, as we did above:

export JAVA_HOME=/usr/lib/j2sdk1.5-sun/
sudo ./sjsas_pe-8_1_02_2005Q2-linux.bin

An install wizard will start, and away you go. Note that you can install without sudo, in which case the app server will be installed by default in your home directory, rather than in /opt.

To make the app server available from NetBeans, fire up NetBeans, go to Tools/Server Manager and Add… your server. Almost all of the server-related tasks you'll want to do can be done from within the IDE now, including – of course – deploying and debugging applications.

Other optional extras:

If desired, change the memory available to NetBeans: edit {install directory}/etc/netbeans.conf, and change the line starting with "netbeans_default_options" to have the desired setting for min/max heap size. For example, on my 2GB laptop, I specify:
"-J-Xms128m -J-Xmx256m"
If you're interested in development for mobile devices, Download and install the mobility pack: for this beta, the file to download is netbeans_mobility-5_0-beta-linux.bin. Just run it, and the next time you run the IDE, you'll be able to create mobile projects. Lukas is on the mobility team, and has interesting and relevant blog entries
A plugin you'll definitely want to install is the profiler: profiler-m9-linux.bin. Again, just run this to install the latest in profiling capabilities .
Finally, by default, NetBeans uses the Metal look and feel on Linux. If you're a keen Ubuntu Gnome person, you might try the GTK L&F, by adding "-J-DuseGtk=true" (without the quotes) to the "netbeans_default_options" line in {install directory}/etc/netbeans.conf. So this line on my system now looks like:
netbeans_default_options=”-J-Xms128m -J-Xmx256m -J-XX:PermSize=32m -J-XX:MaxPermSize=96m -J-ea -J-DuseGtk=true”
Note that this is unsupported, but hey, so is running NetBeans on Ubuntu. One glitch I've found is that with the default "Human" theme in Ubuntu, menu items in Java using this L&F, when selected, disappear (or turn white on white, dunno which). My workaround, until I find out where the problem is, is to change the Ubuntu theme to Glider, which seems to work fine.

CTRL-A to start symbios adapter setup

2005-12-29T13:58:00.000+08:00

and Ctrl -c to set it up

Tomcat problems?

2005-12-26T16:37:00.000+08:00

org.apache.jasper.runtime.PageContextImpl release
WARNING: Internal error flushing the buffer in release()

How do I make Tomcat startup faster

2005-12-23T13:20:00.000+08:00

These are placeholders until more detail is entered ...

Get a faster computer

Make sure your code is not doing slow things. (Use a profiler)?

Remove any jar files you don't need. When searching for classes every JAR file needs to be examined to find the needed class. Also during webapp startup, jar files are searched for TLD files. If the jar file is not there - there is nothing to search.

Remove any webapps you don't need. (So remove the all the webapps installed with tomcat)

Tweak memory parameters - Google is your friend.

Trim the config files as much as possible. XML parsing is not cheap. The less there is to parse - the faster things will go.

I'm starting to look for free email with more than 10 M attachment

2005-12-21T20:51:00.000+08:00

or I have to use split big file into pieces

NetBeans 5.0 Beta Tutorials, Guides, and Articles

2005-12-21T11:45:00.000+08:00

Give you a quick tour of the IDE's support for various technologies by guiding you through the creation of simple applications. These documents generally takes less than 20 minutes to complete

J2EE installation problem in AMD64

2005-12-21T11:32:00.000+08:00

There is a problem to install Java application server in X86-64 linux. Hope there is a solution

Special Report from Nature--Internet encyclopaedias go head to head

2005-12-20T10:08:00.000+08:00

Nature
Published online: 14 December 2005; doi:10.1038/438900a
Internet encyclopaedias go head to head
Jimmy Wales' Wikipedia comes close to Britannica in terms of the accuracy of its science entries, a Nature investigation finds.

Jim Giles

Jimmy Wales started the Internet-based Wikipedia. Click here to hear him talk about our article, and click here to see our list of peer-reviewed encyclopedia entries.
AP PHOTO/M. PROBST
One of the extraordinary stories of the Internet age is that of Wikipedia, a free online encyclopaedia that anyone can edit. This radical and rapidly growing publication, which includes close to 4 million entries, is now a much-used resource. But it is also controversial: if anyone can edit entries, how do users know if Wikipedia is as accurate as established sources such as Encyclopaedia Britannica?

Several recent cases have highlighted the potential problems. One article was revealed as falsely suggesting that a former assistant to US Senator Robert Kennedy may have been involved in his assassination. And podcasting pioneer Adam Curry has been accused of editing the entry on podcasting to remove references to competitors' work. Curry says he merely thought he was making the entry more accurate.

However, an expert-led investigation carried out by Nature — the first to use peer review to compare Wikipedia and Britannica's coverage of science — suggests that such high-profile examples are the exception rather than the rule.

The exercise revealed numerous errors in both encyclopaedias, but among 42 entries tested, the difference in accuracy was not particularly great: the average science entry in Wikipedia contained around four inaccuracies; Britannica, about three.

Considering how Wikipedia articles are written, that result might seem surprising. A solar physicist could, for example, work on the entry on the Sun, but would have the same status as a contributor without an academic background. Disputes about content are usually resolved by discussion among users.

But Jimmy Wales, co-founder of Wikipedia and president of the encyclopaedia's parent organization, the Wikimedia Foundation of St Petersburg, Florida, says the finding shows the potential of Wikipedia. "I'm pleased," he says. "Our goal is to get to Britannica quality, or better."

Wikipedia is growing fast. The encyclopaedia has added 3.7 million articles in 200 languages since it was founded in 2001. The English version has more than 45,000 registered users, and added about 1,500 new articles every day of October 2005. Wikipedia has become the 37th most visited website, according to Alexa, a web ranking service.

But critics have raised concerns about the site's increasing influence, questioning whether multiple, unpaid editors can match paid professionals for accuracy. Writing in the online magazine TCS last year, former Britannica editor Robert McHenry declared one Wikipedia entry — on US founding father Alexander Hamilton — as "what might be expected of a high-school student". Opening up the editing process to all, regardless of expertise, means that reliability can never be ensured, he concluded.

Yet Nature's investigation suggests that Britannica's advantage may not be great, at least when it comes to science entries. In the study, entries were chosen from the websites of Wikipedia and Encyclopaedia Britannica on a broad range of scientific disciplines and sent to a relevant expert for peer review. Each reviewer examined the entry on a single subject from the two encyclopaedias; they were not told which article came from which encyclopaedia. A total of 42 usable reviews were returned out of 50 sent out, and were then examined by Nature's news team.

Only eight serious errors, such as misinterpretations of important concepts, were detected in the pairs of articles reviewed, four from each encyclopaedia. But reviewers also found many factual errors, omissions or misleading statements: 162 and 123 in Wikipedia and Britannica, respectively.

Kurt Jansson (left), president of Wikimedia Deutschland, displays a list of 10,000 Wikipedia authors; Wikipedia's entry on global warming has been a source of contention for its contributors.
D. I. FRANKE/WIKIMEDIA FDN
Editors at Britannica would not discuss the findings, but say their own studies of Wikipedia have uncovered numerous flaws. "We have nothing against Wikipedia," says Tom Panelas, director of corporate communications at the company's headquarters in Chicago. "But it is not the case that errors creep in on an occasional basis or that a couple of articles are poorly written. There are lots of articles in that condition. They need a good editor."

Several Nature reviewers agreed with Panelas' point on readability, commenting that the Wikipedia article they reviewed was poorly structured and confusing. This criticism is common among information scientists, who also point to other problems with article quality, such as undue prominence given to controversial scientific theories. But Michael Twidale, an information scientist at the University of Illinois at Urbana-Champaign, says that Wikipedia's strongest suit is the speed at which it can updated, a factor not considered by Nature's reviewers.

"People will find it shocking to see how many errors there are in Britannica," Twidale adds. "Print encyclopaedias are often set up as the gold standards of information quality against which the failings of faster or cheaper resources can be compared. These findings remind us that we have an 18-carat standard, not a 24-carat one."

The most error-strewn article, that on Dmitry Mendeleev, co-creator of the periodic table, illustrates this. Michael Gordin, a science historian at Princeton University who wrote a 2004 book on Mendeleev, identified 19 errors in Wikipedia and 8 in Britannica. These range from minor mistakes, such as describing Mendeleev as the 14th child in his family when he was the 13th, to more significant inaccuracies. Wikipedia, for example, incorrectly describes how Mendeleev's work relates to that of British chemist John Dalton. "Who wrote this stuff?" asked another reviewer. "Do they bother to check with experts?"

But to improve Wikipedia, Wales is not so much interested in checking articles with experts as getting them to write the articles in the first place.

As well as comparing the two encyclopaedias, Nature surveyed more than 1,000 Nature authors and found that although more than 70% had heard of Wikipedia and 17% of those consulted it on a weekly basis, less than 10% help to update it. The steady trickle of scientists who have contributed to articles describe the experience as rewarding, if occasionally frustrating (see 'Challenges of being a Wikipedian').

Greater involvement by scientists would lead to a "multiplier effect", says Wales. Most entries are edited by enthusiasts, and the addition of a researcher can boost article quality hugely. "Experts can help write specifics in a nuanced way," he says.

Wales also plans to introduce a 'stable' version of each entry. Once an article reaches a specific quality threshold it will be tagged as stable. Further edits will be made to a separate 'live' version that would replace the stable version when deemed to be a significant improvement. One method for determining that threshold, where users rate article quality, will be trialled early next year.

To see a list of the peer-reviewed encyclopedia articles, click here.

Additional research by Declan Butler, Jenny Hogan, Michael Hopkin, Mark Peplow and Tom Simonite.

Article brought to you by: Nature

Sun Java Studio Creator Tutorials

2005-12-19T21:39:00.000+08:00

Sun Java Studio Creator � Tutorials

Search Engines Registration

2005-12-19T19:42:00.000+08:00

http://www.trafficzap.com/searchsubmit.php
which should be updated monthly

Ripple's Blog

2005-12-19T18:25:00.000+08:00

Ripple's Blog

NetBeans IDE Look & Feel Competition

2005-12-19T16:34:00.000+08:00

Have you customised your NetBeans IDE Look & Feel ? Are you using a non-standard Look&Feel, or are you using the Substance plug-in to change NetBeans' look ? Have you tweaked the L&F in some other way ?
If so, share your customisation and you could win one of three iPod Nanos, 10 USB flash memory sticks or 30 NetBeans T-Shirts! More competition details please go to http://www.netbeans.org/competition/look-and-feel.html.

CVS server on Ubuntu

2005-12-19T14:54:00.000+08:00

Install CVS files:
sudo apt-get install cvs

Install the CVS server:
sudo apt-get install cvsd

When prompted in the cvsd installation process for Repository, type in “/cvsrepo”.

Now that the cvsd installation in complete goto /var/lib/cvsd

or seeking for a change(or if there is a new version of cvs updated):

sudo cvsd-buildroot /var/lib/cvsd

If the folder cvsrepo does not exist, then create it ..
sudo mkdir cvsrepo

sudo chown -R cvsd:cvsd cvsrepo

and then initilize the repository
sudo cvs -d /var/lib/cvsd/cvsrepo init

create a user and password

sudo cvsd-passwd /var/lib/cvsd/cvsrepo +username

sudo vi //var/lib/cvsd/cvsrepo/CVSROOT/config

Change "SystemAuto=no"

Test

cvs -d :pserver:username@localhost:/cvsrepo login

cvs -d :pserver:username@localhost:/cvsrepo checkout .

Split and Reassemble files [2]

2005-12-19T14:48:00.000+08:00

Recreating the file (cat)
Recreation of the original file is even easier. For this, we will be using 'cat'. See the man pages for more information on 'cat' (man cat).
The command that we will use is
cat x* > your original file

(I'd recommend that backup your file before any operation)

Split and Reassemble files [1]

2005-12-19T14:42:00.000+08:00

here always comes a time, where you wish that file was only a few kilobytes/megabytes smaller. Wether it be so it can fit onto your floppy disk, CDR etc, or so you can meet the attachment limit on an e-mail server. This isn't really a command that you would use everyday, but it might come in handy.
Splitting the file (split)
Now, for this example I will use the file hello.mp3 which is more than 10M

The command that we will use is $ split -b 2048k hello.mp3

$ split -b 2048k hello.mp3
a list of files will be generated
xaa
xab
xac
xad
...
OK, what we have now done is told 'split' to use bytes per output file (-b). For more information and arguments on split, view the man pages (man split).
These files now meet the e-mail server's attachment limit and can be sent.

Compiling Issue on AMD64

2005-12-19T14:20:00.000+08:00

Recently We have a New AMD64 HP workstation Opteron with 4G memory. It's also my 1st experience to do my jobs under 64bit envirment.

However there are still many issues on it. One of them is the compiling problem for many software, after ./configure, it complaines that the systems is probelmatic: Invalid configuration `x86_64-unknown-linux-gnu': machine `x86_64-unknown' not recognized.

Many of this problem have been resolved by giving a new config.sub that is aware of x86_64.
for example:

cp /usr/share/automake-1.7/config.sub . ./configure

Welcome to visit my Blog

2005-12-19T10:45:00.000+08:00

Welcome to this Blog and welcome to my life. Here I aim to provide information about my uderstanding and experience in programming, database managment, sports, music and my many other favoriate interest.
Hope you like it.