Assignment Cover Sheet Internal

brewerobstructionΤεχνίτη Νοημοσύνη και Ρομποτική

7 Νοε 2013 (πριν από 3 χρόνια και 9 μήνες)

106 εμφανίσεις

Page
1

of
20



UNIVERSITY OF SOUTH AUSTRALIA


Assignment Cover Sheet



Internal



An Assignment cover sheet needs to be
included with

each

assignment.

Please
complete

all details clearly
.


If you are submitting the assignment on pape
r, please staple this sheet to the front of each assignment. If you are
submitting the assignment online, please ensure this cover sheet is

included

at the start of your document. (This is
preferable to a separate attachment.)


Please check your
Course Inf
ormation Booklet

or contact your School Office for assignment submission locations.


Name: Phirun Son

Student
ID


1

0

0

0

6

4

5

1

7

Email: sonpy003@students.unisa.edu.au

Course code and title:

CIS Research Methods INFT 4017

School: CIS

Program code:
LHIS

Course Coordinator
:

Prof. Paul Swatman

Tutor: Prof. Paul Swatman

Day, Time, Location of Tutorial/Practical: Thursday 9
-
12

Assignment number:


Due date:

14 June
2009

Assignment topic as stated in
Course
I
nformation

Booklet:

Research Proposal


Further Information:
(e.g.

state if extension was granted

and attach evidence of approval, Revised Submission Date
)




I declare that the work contained in this assignment is my own, except where acknowledgement of sources is made.

I

authorise

the University to test any work submitted by me, using text comparison software, for instances of plagiarism. I
understand this will involve the University or its contractor copying my work and storing it on a database to be used in
future to test work su
bmitted by others.

I understand that I can obtain further information on this matter at

http://www.unisanet.unisa.edu.au/learningconnection/student/studying/integrity.asp

Note:
The attachment of this statement on any electronically submitted assignments
will be deemed to have the same
authority as a signed statement.


Signed:

Phirun Son

Date:

14 June

2009


Date received from student



Assessment/grade

Assessed by:


Recorded
:

Dispatched

(if applicable):



Page
2

of
20


University of South Australia


School of
Computer and Information Science

Bachelor of Information Science

(Advanced Computer and Information Science)


CIS Research Methods

Research Proposal


Modelling
Micro
array Data with Bayesian
Networks Using Structure Learning


Student: Phirun Son

Student ID:

100064517

Supervisor: Dr Lin Liu



Page
3

of
20


Table of Contents


Abstract

................................
................................
................................
................................
...................

4

Introduction

................................
................................
................................
................................
............

5

Background

................................
................................
................................
................................
.........

5

Research Question

................................
................................
................................
..............................

6

Purpose and Limitations

................................
................................
................................
.....................

7

Literature Review

................................
................................
................................
................................
....

9

DNA Microarray

................................
................................
................................
................................
..

9

Bayesian Networks

................................
................................
................................
............................

10

Struct
ure Learning

................................
................................
................................
............................

11

Summary

................................
................................
................................
................................
...........

12

Research Method

................................
................................
................................
................................
..

13

Platform

................................
................................
................................
................................
............

14

Expected Outcome

................................
................................
................................
............................

15

References

................................
................................
................................
................................
............

17




Page
4

of
20


Abstract


Gene microarray
s

are

a technology that allows the measurement of
several thousand genes’
expression levels. It is captured as an array of
thousands of small dots
,

which represents
a snapshot
of the expression levels of the genes.

This allows the acquisition of large amounts of data with
regards to
what genes are responsible for causing

(Shalon
et al,
1996)
. However, as this data
become
s

more

prevalent, due to the complexity of the data, it would be required that some
automation is use
d in order to analyse the data.


The challenge of determining methods of analysing the vast amounts of data is important as it may
allow the discovery of causal
dependencies and relations of different genes, and
maybe
eventually,
model

the
overall structure of the gene network
. There have been
past
attempts to
model
microarray data into causal networks, including the use of Bayesian Networks. The
re are

character
is
tics of Bayesian Networks that are desirable when working with microarray data as it can
attempt to learn network structure from data that is incomplete or noisy, as microarray data is

(Heckerman, 1998)
. The difficulty is determining the correctness of the

structures found, but even
partially correct graphs can assist in further research being performed to validate the precision of
the results.


In order to determine the correctness of structures found by the learning of data, what can be done
is take exist
ing known structures of networks and use sample data to try to recreate the structure.
This can help validate the accuracy of the algorithms used to determine the structure, or find flaws
in the algorithm. By testing and comparing various proposed algorith
ms, an optimal one may be
found which can be applied on new or existing gene data.



Page
5

of
20


Introduction


Background


Genetic research is a major field of research in the bioinformatics field of recent. Specifically, a large
amount of attention has been growing i
n relation to understanding the gene regulatory
networks,
also known as gene expression networks, which govern the way our bodies control cellular
functions.


Deoxyribonucleic acid

(DNA) is the template that our bodies use to create our cell structures. It

is a
form of nucleic acid which contain
s

genetic instructions, which it relays to other components of
our
genetic makeup
for the purpose of protein creation. This is done by creating messenger ribonucleic
acid (mRNA), whose purpose is to transcribe DNA se
quence information into RNA sequence
information.

Ribonucleic acid (RNA) contains the information and components needed to create
cellular proteins

(Knox
et al,
2001)
.


The process of transcribing this information is known as gene expression, or gene regulation. The
levels in which these genes are expressed
are

able to be measured in a variety of ways, but a certain
method
that has allowed an unprecedented scope of data i
s the use of microarrays

(Shalon
et al,
1996).


A gene microarray is a technology which is able to capture thousands of genes’ expression levels at a
time, giving a huge amount of data. This is done by having a solid surface, such as a glass or silicon
chi
p, attached with thousands of microscopic spots. These spots are known as probes, and are
generally a short section of a gene, which can be processed using chemicals such as fluorophores in
order to determine the level of expression of the gene.

The expres
sion levels of these targeted genes
are generally captured over a period of time, in which the genes are exposed to some form of
stimuli, such as a drug treatment, giving the progression of expression levels as a reaction to the
stimuli

(Shalon
et al,
1996
).


Page
6

of
20


It

can be ascertained that this method of data capture of gene expression levels creates a huge
amount of data to interpret. There have been many attempts at realising this data into a formal
structure, so that causal relationships between genes can be

found, including the use of Boolean
network

and

linear structures
, a large focus is put on Bayesian networks as a means of constructing
the gene regulatory network. Bayesian networks have several desirable features which will be
described later, which giv
e it an advantage in modelling the data formed by microarray experiments.



Research Question


There have been many novel proposed techniques which aim to ascertain a gene regulatory network
by evaluating microarray data, however no one technique is foolpr
oof in creating an accurate model,
and little information is known about gene structures to validate whether the results of these
techniques are correct. The way in which we can move forward in this area is to prove that certain
techniques are viable with
existing known structures, by taking these known structures and trying to
recreate them using only data, and seeing whether they are sufficiently accurate toward the original
data structure.


This paper will aim to ex
amine

the following questions:

Are Bayesian
network structures suitable for modelling gene regulatory
networks?

This question will determine whether Bayesian networks are truly a viable method of
determining the structure of a gene network.


Which methods of structure learning are best
used to create an accurate model of a gene
network?

This question will help to identify the most promising of current techniques used in Bayesian
networks for use in determining a network structure from data.


Page
7

of
20


Are these methods viable for use in helping to

determine the network structure of an
unknown network?

This question will illustrate the practical use of using Bayesian network to discover actual
gene networks, and help validate that the techniques used are feasible for real world use.


In order to ans
wer these questions, a testing environment will be setup in MATLAB, an established
programming environment used in scientific and mathematical research.



Purpose

and Limitations


Gene research is an important element in discovering more and more about how

our bodies
function. It can help in founding new medical research and in developing treatments and cures for
diseases. If we could better understand the way that our genes interact with one another to produce
the building blocks of our bodies, it is possi
ble to further research into
genetic science.


This can be achieved by using the data collected from microarray experiments and trying to estimate
a causal network of which genes interact and are dependent on other genes. This is a large problem
as, althou
gh there have been established methods of modelling data, the challenges of working with
gene data are many, the most daunting of which is the sheer number of genes present in our DNA
makeup, let alone an experiment which only covers a few thousand of thos
e.


As stated earlier, there
is already a good deal

of novel approaches towards modelling gene
networks. Most of these techniques take existing algorithms in the field of structure learning of a
Bayesian network, and modify them in order to create more acc
urate results. The scope of creating
such a modification such as these is too great to be done
within the time limitation of this research,
therefore this research will focus on taking these known existing techniques and applying them in
order to compare a
nd validate which ones are most viable.

Page
8

of
20



The work done will contribute to
the knowledge base of which techniques of structure learning are
most suitable, and whether certain techniques should be pursued further. It may help to guide
future researchers in d
etermining which techniques would be best suited to focus upon and expand
the ideas of that technique, and which ones should possibly be ignored due to the inaccuracy or
infeasibility of it.



Page
9

of
20


Literature Review


DNA Microarray


DNA microarray technology is a fairly recent technology, being incepted in the mid 1990’s.
Along
with allowing an unprecedented amount of genes to be monitored at once, microarray technology
solved problems that earlier methods of monitoring gene expressio
n levels faced, such as in blotting.
For instance, blotting required
the use of porous membranes on which the gene probes were
attached. This limited the scope of genes that could be measured due to the requirement of needing
radioactive, chemiluminescent
or colormetric detection methods. These methods cause the probe
readings to scatter and disperse. This is not the case with microarray technology, which can be
applied on a glass surface, and uses fluorescent detection methods, resulting in lower backgroun
d
interference and greater probe density

(Shalon
et al,
1996)
.

This not only allows for more genes to
be measured at once, but increases consistency as they are measured during the same experiment
rather than several, which would be later normalised.


Micr
oarray technology is not the only technology in recent years to be discovered; there have been
other types of methods that have emerged, such as serial analysis of gene expression (SAGE)
(Velculescu
et al,
1995),
however microarray technology has taken off

the most due to being
relatively easy to use, due to not requiring radioactive materials, and therefore specialised labs
(Russo
et al,
2003).
Another advantage is that microarray technology is relatively cheap, and there
are various types of microarray te
chnology with incremental costs, allowing a range of firms of
various sizes to afford to perform some kind of microarray
-
based technology (Granjeaud
et al,
1999).


Due to the popularity and relative ease of microarray technology, there is a huge influx of
microarray
data of which there is little ability to process for valuable information (Granjeaud
et al,
1999). To
compound on this, there has been little in the way of standardising the format of microarray data.
There are many different manufacturers and p
rocedures used to collect the gene data, so naturally
there is variation in how the final data is represented. There have been attempts to standardise the
format of microarray data to relieve this problem, with the most widely accepted being ‘Minimum
Infor
mation About a Microarray Experiment’ (MIAME) developed by the Microarray and Gene
Page
10

of
20


Expression Data Society (MGED) (Brazma
et al,
2001).

This standardisation has allowed repositories
of microarray data to be formed, housing free gene expression data to the
public. Such databases
include ArrayExpress from the European Bioinformatics Institute (
Brazma
et al,
2003
) and Gene
Expression Omnibus (GEO) from the
National Center for Biotechnology Information

(Edgar
et al,
2002)
, both of which contain an abundance of
freely available microarray data
.


Although there is an abundance of data available in a standard format, there is still the problem of
modelling this data. This is a difficult challenge due to the nature of gene microarray data, but there
have been many a
ttempts put forth.


Bayesian Networks


There have been different ideas as to what model would be best suited to model gene networks,
such as neural networks (Xu
et al,
2007), linear equations (Gebert
et al,
2007) and Boolean networks
(Shmulevich
et al,
200
2), but the model that seem to have the most promise are Bayesian networks
.


Bayesian networks, also known as belief networks, represent a set of variables in the form of nodes
on a directed acyclic graph (DAG). It maps the conditional independencies of these variables.
According to Heckerman (1998), Bayesian networks bring us four

advantages as a data modelling
tool, three of which are directly beneficial to working with gene microarray data.
Firstly, Bayesian
networks are able to handle incomplete or noisy data, which is a common trait of microarray
experiments due to the nature o
f how the data is captured. Secondly, Bayesian networks are able to
ascertain causal relationships through conditional independencies, allowing the modelling of
relationships between genes. The last advantage is that Bayesian networks are able to incorpora
te
existing knowledge, or pre
-
known data into its learning, allowing more accurate results by using
what we already know. These points are re
-
iterated in many papers as reasoning of why Bayesian
networks are a viable solution to modelling microarray data (
Chen
et al,
2006
;
Spirtes

et al

2001;
Wang
et al,
2007; Yavari
et al,
2008).


Page
11

of
20


The draw of using Bayesian networks for purposes of modelling gene networks are that Bayesian
networks are able to learn the structure of a DAG by observing data. There are many
different
techniques used to do this, however the challenge is in finding a suitable technique for use on gene
data, of which there are many difficulties.


Structure Learning


Structure learning is the act of finding a plausible structure for a graph based

on data input.
However, it has been proven that this is an NP
-
Hard problem (Chickering, 1996), and therefore any
learning algorithm that would be appropriate for use on such a large dataset such as microarray data
would require
some form of modification f
or it to be feasible.


It is explained by Spirtes
et al

(
2001) that finding the most appropriate DAG from sample data is
large problem as the number of possible DAGs grows super
-
exponentially with the number of nodes
present. With gene data, this number is

typically in the thousands, if not more.


The K2 algorithm, a widely used technique for structure learning is one such standard learning
algorithm which uses heuristics in order to limit the search space of
DAGs. It requires a node
ordering, that is the
variables given to the algorithm must be from ancestor to descendant, and
works on the principle of assuming a node has no parents. It then incrementally adds parent nodes
and calculates probability until no parent node is found to be beneficial (Cooper &
Herskovits, 1992).


By noting the method that the K2 algorithm operates, it is easy to deduce that the ordering of nodes
is a large factor in determining the final result found. This poses the difficulty of not knowing prior
ordering in a network such as a gene network.
There

have been attempts at modifying the K2
algorithm in order to create a more feasible form, such as an effort by Numata
et al
(2007) to learn
more appropriate orderings by observing a number of distributed orders, and
narrowing down order
through observatio
n. Another effort was made by Chen
et al
(2006) which split the task of searching
by first creating an undirected graph using mutual information. The graph is then processed using
the reduced search space in order to assign directions to the undirected edg
es.

Page
12

of
20



Other means of simplifying the problem have been to use hybrid techniques, combining different
methods together in order to try getting better results. For instance, the use of clustering, by which
genes with similar expression profiles are group
ed to
gether in order to reduce the sample space.
This allows conventional algorithms which can run with the smaller number of variables to be used.
These network clusters can then be used to incrementally increase the scope of the network (Yavari
et al,
2008; Z
ainudin & Deris, 2008).


There have also been efforts in which new algorithms are being proposed to deal with the nature of
gene data. Pena
et al
(2005) proposed a technique whereby a seed gene (arbitrary) is used as a base
upon where
a certain number of
d
ependent genes are found for that gene, and iteratively increase
the
number of dependant genes to look for.


Summary


It can be seen that there has been a lot of effort put forward in pursuit of finding a reasonable
method for modelling data from gene
microarray experiments. The problem is still being pursued,
and there has not been a completely optimal solution to the problem as of yet. In order to further
research how this task can be achieved, we need to build upon the knowledge that is afforded to u
s
by prior efforts.




Page
13

of
20


Research Method


Although there are a fair amount of established techniques used to model data into a Bayesian
network, they are not all feasible to perform when dealing with gene data, whether for time
limitation, memory limitation,

or accuracy. This is why there have been novel, innovative approaches
to adjusting and modifying some of these established techniques in order to give better results in
regards to
accurately modelling gene data
.


Though there are many approaches to this p
roblem, it would not be possible to test every one of
them. Therefore, it would be required to first read through and select those which seem to be the
most promising in order to further look into them and see whether they are as useful as they are
proclai
med.


The criteria for choosing which approaches will be selected are as follows:


The technique must be applicable to a general Bayesian network structure.

There are other forms of structures that are proposed to use for creating a gene network
structure,

however the focus of this research is to only consider Bayesian networks.

The technique must be able to be tested on a standalone machine.

The machine that will be used for testing is a standalone
computer;

therefore it does not
have the capabilities of
using networked or clustered computer to do any processing. This
limits the techniques that can be tested to those that do not
require

these advance
processing requirements.



Page
14

of
20


Platform


In order to test the techniques chosen, we will require a platform in
which the various algorithms
can be run and tested on data with equal footing. Therefore a system where data from gene
microarray experiments can be imported into is needed, as well as a way to represent Bayesian
network structures and the capability to pe
rform the intended algorithms on them.


For this purpose, the environment of MATLAB has been chosen

(Mathworks, 2009)
. MATLAB is a
well established environment of which many mathematical and scientific functions are already pre
-
programmed, and it has the a
bility to extend its capabilities through toolboxes. In particular, the
MATLAB environment is fairly easy to use as it caters towards those who are not usually well versed
in the computer science field, so even complex activities can be performed with ease
.


In terms of extendable toolboxes, there is a free, open source toolbox known as the Bayesian
Network Toolbox (BNT)

(Murphy, 200
1
)
. As the name implies, the toolbox extends MATLAB to
include the data structures and functions to allow Bayesian Networks to

be represented and
manipulated. This simplifies the process of needing to create a representation of Bayesian networks
ourselves, and also comes with many
convenient functions which can be used to help in processing
data into Bayesian networks. This allow
s for more time to focus on implementation and testing of
algorithms.


Along with the BNT, there is another extension called the Structure Learning Package

(SLP)

(Leray &
Francois, 2004)
, which extends the BNT even more by adding new functions in order to
perform
structure learning with various common structure learning algorithms, such as Markov Chain

Monte
Carlo (MCMC), PC, K2 algorithm, Maximum Weight Spanning Tree (MWST), Greedy search, etc.
This
gives a good basic structure for beginning to test struct
ure learning with, and as most of the novel
proposed techniques are modifications of these learning algorithms, it gives an established base in
which is proven to be correct and will not need testing of those parts. The SLP is also a free open
source toolb
ox extension for MATLAB.


Page
15

of
20


As both toolboxes that have been chosen are open source projects, they are available for
modification, and so are suitable to use for extending them to include the other algorithms that are
chosen to be tested.


Alongside the MATL
AB environment, there has been experimentation with Hugin Researcher

(Hugin
Expert, 2009)
. It is a powerful tool which has built in functions for networks and has its own
structure learning algorithms, however it is not easily extendable and therefore it i
s not certain how
far I will use the tool.



Expected Outcome


The outcome expected to be produced as a result of this research is
an analysis of the techniques
that I have chosen to test. The analysis would have statistics on several points of interest, s
uch as
accuracy

or efficiency, and special considerations when using these various techniques. For instance,
some techniques may only be suitable if certain conditions are met, or some may require additional
data in order to perform properly.


This data ma
y be useful to other researchers who are delving into the field of gene regulatory
networks as it can help them glance at a study of which techniques are most practical in determining
causal network structures.


Another outcome will be to have a practical
example of how the algorithms may or may not have
helped in analysing a new data set which has not been analysed before. It can show whether these
techniques have helped in estimating an accurate causal network of a real life data set. This part is
depende
nt on another researcher, however, and the results may not be known until after this
research project is over.


Page
16

of
20


This outcome will show that the techniques used are practical and can be use to perform analysis on
real world data.
Scientists may be able to u
se this research to validate that the estimate gene
networks are reasonable enough to use as a base to test for true causal networks of genes.


One more outcome that may be produced is another extension to the Bayesian Network Toolkit.
Depending on the sig
nificance of how much is put into modifying and implementation of the chosen
test algorithms from external papers, the modifications may be released so that other users of the
toolkit can choose to download these extra algorithms to experiment with. Any ex
tension released
would be put under the GNU Public Licence (GPL).



Page
17

of
20


References


Brazma, A, Hingamp, P, Quackenbush, J, Sherlock, G, Spellman, P, Stoeckert, C, Aach, J, Ansorge, W,
Ball, C & Causton, H 2001, 'Minimum information about a microarray
experiment (MIAME)

toward
standards for microarray data', Nature genetics, vol. 29, pp. 365
-
371.


Brazma, A, Parkinson, H, Sarkans, U, Shojatalab, M, Vilo, J, Abeygunawardena, N, Holloway, E,
Kapushesky, M, Kemmeren, P & Lara, G 2003, 'ArrayExpress
--
a publ
ic repository for microarray gene
expression data at the EBI', Nucleic Acids Research, vol. 31, no. 1, p. 68.


Chen, X, Anantha, G & Wang, X 2006, 'An effective structure learning method for constructing gene
networks', Bioinformatics, vol. 22, no. 11, pp.

1367
-
1374.


Chickering, D 1996, 'Learning Bayesian networks is NP
-
complete', Learning from data: Artificial
intelligence and statistics v, pp. 121

130.


Cooper, G & Herskovits, E 1992, 'A Bayesian method for the induction of probabilistic networks from
da
ta', Machine learning, vol. 9, no. 4, pp. 309
-
347.


Edgar, R, Domrachev, M & Lash, A 2002, 'Gene Expression Omnibus: NCBI gene expression and
hybridization array data repository', Nucleic acids research, vol. 30, no. 1, p. 207.


Gebert, J, Radde, N & Weber
, G 2007, 'Modeling gene regulatory networks with piecewise linear
differential equations', European Journal of Operational Research, vol. 181, no. 3, pp. 1148
-
1165.


Granjeaud, S, Bertucci, F & Jordan, B 1999, 'Expression profiling: DNA arrays in many gui
ses',
Bioessays, vol. 21, no. 9, pp. 781
-
790.

Page
18

of
20



Heckerman, D 1998, 'A tutorial on learning with Bayesian networks', NATO ASI SERIES D
BEHAVIOURAL AND SOCIAL SCIENCES, vol. 89, pp. 301
-
354.


HUGIN EXPERT 2009,

Hugin Researcher 7.0,
<http://www.hugin.com/Pro
ducts_Services/Products/Academic/Researcher/>.


Knox, L & Evans, S 2001, Biology (2nd edn).', McGraw
-
Hill Book Company: Sydney.


Leray, P & Francois, O 2004, 'BNT structure learning package: Documentation and experiments',
Laboratoire PSI, Tech. Rep.


Mathworks 2009, MATLAB 2008b, <http://www.mathworks.com.au/products/matlab/>.


Murphy, K 2001, 'The bayes net toolbox for matlab', Computing science and statistics.


Numata, K, Imoto, S & Miyano, S 2007, 'A Structure Learning Algorithm for
Inference of Gene
Networks from Microarray Gene Expression Data Using Bayesian Networks', paper presented at the
Bioinformatics and Bioengineering, 2007. BIBE 2007. Proceedings of the 7th IEEE International
Conference on.


Pena, JM, Bjorkegren, J & Tegner,

J 2005, 'Growing Bayesian network models of gene networks from
seed genes', Bioinformatics, vol. 21, no. suppl_2, September 1, 2005, pp. ii224
-
229.


Russo, G, Zegar, C & Giordano, A 2003, 'Advantages and limitations of microarray technology in
human ca
ncer', Oncogene, vol. 22, no. 42, pp. 6497
-
6507.

Page
19

of
20



Shalon, D, Smith, SJ & Brown, PO 1996, 'A DNA microarray system for analyzing complex DNA
samples using two
-
color fluorescent probe hybridization', Genome Res., vol. 6, no. 7, July 1, 1996,
pp. 639
-
645.


Sh
mulevich, I, Dougherty, E & Zhang, W 2002, 'From Boolean to probabilistic Boolean networks as
models of genetic regulatory networks', Proceedings of the IEEE, vol. 90, no. 11, pp. 1778
-
1792.


Spirtes, P, Glymour, C, Scheines, R, Kauffman, S, Aimale, V & Wi
mberly, F 2001, 'Constructing
Bayesian network models of gene expression networks from microarray data'.


Velculescu, V, Zhang, L, Vogelstein, B & Kinzler, K 1995, 'Serial analysis of gene expression', Science,
vol. 270, no. 5235, p. 484.


Wang, M, Chen, Z

& Cloutier, S 2007, 'A hybrid Bayesian network learning method for constructing
gene networks', Computational Biology and Chemistry, vol. 31, no. 5
-
6, pp. 361
-
372.


Xu, R, Wunsch, D & Frank, R 2007, 'Inference of genetic regulatory networks with recurrent

neural
network models using particle swarm optimization', IEEE/ACM Transactions on Computational
Biology and Bioinformatics, vol. 4, no. 4, pp. 681
-
692.


Yavari, F, Towhidkhah, F, Gharibzadeh, S, Khanteymoori, AR & Homayounpour, MM 2008, 'Modeling
Large
-
S
cale Gene Regulatory Networks using Gene Ontology
-
Based Clustering and Dynamic Bayesian
Networks', paper presented at the Bioinformatics and Biomedical Engineering, 2008. ICBBE 2008.
The 2nd International Conference on.


Page
20

of
20


Zainudin, S & Deris, S 2008, 'Combi
ning Clustering and Bayesian Network for Gene Network
Inference', paper presented at the Intelligent Systems Design and Applications, 2008. ISDA '08.
Eighth International Conference on.