Fill - Free fillable Original Article Clinical Proteomics Copyright 2006 Humana Press PDF form

Original Article

Data Mining in Proteomic Mass Spectrometry

Asha Thomas,

Georgia D.Tourassi,

Adel S. Elmaghraby,

Roland Valdes Jr.,

and Saeed A. Jortani,

Department of Computer Engineering and Computer Science, University of Louisville, Louisville, KY;

Digital Advanced Imaging Laboratories, Department of Radiology, Duke University Medical Center,

Durham, NC; and

Department of Pathology and Laboratory Medicine, University of Louisville,

Louisville, KY

*Author to whom all correspondence and reprint requests should be addressed:

Saeed A. Jortani, Department of Pathology, University of Louisville, Louisville, KY.

E-mail: sjortani@louisville.edu.

Clinical Proteomics

All rights of any nature whatsoever are reserved.

ISSN 1542-6416/06/02:13–32/$30.00 (Online)

Abstract

Data mining application to proteomic data

from mass spectrometry has gained much

interest in recent years. Advances made in pro-

teomics and mass spectrometry have resulted

in considerable amount of data that cannot be

easily visualized or interpreted. Mass spectral

proteomic datasets are typically high dimen-

sional but with small sample size. Conse-

quently, advanced artificial intelligence and

machine learning algorithms are increasingly

being used for knowledge discovery from such

datasets. Their overall goal is to extract useful

information that leads to the identification of

protein biomarker candidates. Such biomark-

ers could potentially have diagnostic value as

tools for early detection, diagnosis, and prog-

nosis of many diseases. The purpose of this

review is to focus on the current trends in

mining mass spectral proteomic data. Special

emphasis is placed on the critical steps

involved in the analysis of surface-enhanced

laser desorption/ionization mass spectrometry

proteomic data. Examples are drawn from pre-

viously published studies and relevant data

mining terminology and techniques are

explained.

03_Jortani.qxd 11/21/06 3:07 PM Page 13

Clinical Proteomics________________________________________________________________ Volume 2, 2006

14 _______________________________________________________________________________ Thomas et al.

Key Words: Mass spectrometry; data mining;

artificial intelligence; data analysis; SELDI; linear

discriminant analysis; artificial neural networks.

Introduction

As the technology involved in biological

domains like proteomics advances, large vol-

umes of data are continuously generated.

However, the data analysis required is

increasingly overwhelming to the scientists

in these domains. Contrarily, computational

scientists are accustomed to dealing with

large databases from domains such as mar-

keting, financial institutions, and telecommun-

ication. They typically face the challenge by

the use of powerful machine learning and

data mining tools. Such tools are the culmina-

tion of continuous advances made in the areas

of artificial intelligence, statistics, and database

management. Therefore, it is natural that the

proteomics and computational intelligence

domains find their paths crossed with the

common goal of extracting clinically useful

information from the wealth of proteomic

spectral data.

The purpose of this review is to address

some of the key issues involved in the appli-

cation of data mining approaches to proteomic

data from surface-enhanced laser desorption/

ionization mass spectrometry (SELDI-MS). In

addition, terminologies and popular strategies

in mining of mass spectral proteomic data are

described. The article is organized as follows.

First, a brief introduction to data mining and

proteomics is provided. Then, the need for

applying data mining to proteomic data from

SELDI-MS is explained. Thereafter, important

steps in the mining process are outlined and

explained with examples drawn from recent

studies involving SELDI-MS data.

Data Mining in Proteomics

Proteomics

Proteomics is an emerging area in bioinfor-

matics. First coined by Australian scientist

Marc Wilkins in 1994 (1), proteomics is a sci-

ence that deals with the study of proteins and

their interactions in an organism. The major

focus of most proteomic studies is the dis-

covery of the proteins of interest known as

biomarkers (2). The discovery of biomarkers

is potentially valuable for the early detection,

diagnosis, and monitoring of diseases (e.g.,

refs. 3–6). Active and ongoing biomedical

research advances coupled with the discov-

ery of novel and powerful diagnostic pro-

teomic tools have further strengthened the

progress made.

There are several techniques for protein or

peptide profiling using mass spectroscopy.

Two of the major techniques intended for pro-

teomic mass spectrometry are matrix-assisted

laser desorption/ionization (MALDI-MS) (7a)

and its extension, the SELDI-MS (7–8). In both

techniques, the biological matrix is mixed with

the energy absorbing matrix and following

shinning of the laser energy in vacuum envi-

ronment, which is absorbed by the proteins

causing them to get ionized. After application

of an electric field, the ions accelerate through

a flight tube until finally being detected by the

instrument. The major difference between

MALDI and SELDI is the fact that the latter

uses additional chemistries on the surface of

the chips to further isolate the proteins

intended to be analyzed from the other

molecules or proteins in the matrix (7). This

review will focus mainly on issues related to

the data mining of SELDI proteomic data

because many of the recent discoveries of

potential new biomarkers have involved this

technology.

Typically, protein profiles in body fluids

such as serum, urine, or nipple aspirate, and

in some studies tumor tissue, are analyzed

using SELDI mass spectrometry (Ciphergen

Biosystems, Fremont, CA). The output is a

protein expression profile that is often a dataset

containing information of protein peaks and

their heights (intensities) in one or more

03_Jortani.qxd 11/21/06 3:07 PM Page 14

Volume 2, 2006________________________________________________________________ Clinical Proteomics

Data Mining in Proteomic MS ________________________________________________________________ 15

groups of patients (e.g., cancer, benign,

healthy). The main goal of proteomic studies

using mass spectral data is to identify protein

patterns that are capable of reliably differentia-

ting between various groups under study. The

identification of these protein biomarker pat-

terns that relate to a certain pathological state

may aid in the early detection and prognosis

of that disease.

However, identification of these biomarker

patterns from mass spectral data is a challeng-

ing task. Issues such as few samples and large

number of peaks, typical in MS data, have to

be handled by advanced analytical techniques.

Data Mining

In several studies, protein profiles gener-

ated from SELDI-MS have been analyzed

using statistical and data mining methods.

Data mining is an approach used in scientific

and business domains to extract meaningful

and useful information from large and com-

plex sets of data. The process of data mining

is typically iterative (9) wherein previously

unknown and potentially useful information

is discovered by the use of powerful analyt-

ical techniques. The discovery process

involves finding relationships and patterns in

raw data that can be either utilized or assessed

by decision makers and analysts. Steps

involved in mining include data preparation,

feature selection, model development or pat-

tern recognition, and model assessment fol-

lowed by application of developed model to

new cases.

The early use of data mining was in finan-

cial institutions and marketing (10–13). How-

ever, with an explosion in the amount of data

being stored electronically, data mining has

reached across telecommunications (14), retail

(15), manufacturing (16), health care (17),

fraud detection (18), homeland security (19),

and biomedical domains (20).

Data mining methods have evolved from

advancements made in artificial intelligence,

statistics, and database management. There

are various data mining algorithms that are

available today, based on different theoretical

concepts. Some are in the form of decision

trees (DTs), known for their ease of inter-

pretability. Others, like artificial neural net-

works (ANNs), capture complex and nonlinear

relationships in the data. ANNs are considered

very powerful but are less interpretable. The

advantages and disadvantages of each method

when applied to SELDI-MS data are discussed

in the following sections.

Application of Data Mining in SELDI-MS

Proteomic Data

As previously mentioned, the use of mass

spectrometry in proteomics results in a large

number of high-dimensional data, where the

number of features (peaks) is much greater

than the number of samples. Data samples

from SELDI-MS are typically comprised of

hundreds to thousands of protein peaks. Such

a vast number of data cannot be visually ana-

lyzed or handled by ordinary data mining

tools. Tackling such a high-dimensional struc-

ture of a comparatively small-sized data set is

a significantly challenging task. An example of

two typical spectra for serum samples is

depicted in Fig. 1.

In their quest for adequate tools for analyz-

ing the data in hand and retrieving useful

information, scientists in proteomics are

increasingly dependent on advanced data

mining techniques that can tackle issues such

as the curse of dimensionality and limited

data sets. These advanced techniques include

artificial intelligence and machine learning.

Current practices in mining protein mass

spectrometry data from SELDI include the fol-

lowing steps:

1. Data modeling using peaks identified by

preprocessing and feature selection.

2. Careful data sampling to address the small

sample size typical in SELDI-MS data.

3. Performance evaluation of the data models.

03_Jortani.qxd 11/21/06 3:07 PM Page 15

Clinical Proteomics________________________________________________________________ Volume 2, 2006

16 _______________________________________________________________________________ Thomas et al.

The previously listed critical steps have to be

carefully addressed by proteomic researchers to

obtain accurate and robust decision models.

The steps are repeated iteratively and modifi-

cations are made so as to probe and explore

different aspects of the data. Figure 2 outlines

the mining process and the possible methods

that are available under each of them. Each

data-processing step is described in more

detail in the following sections.

Preprocessing

The raw data obtained from SELDI-MS is

noisy. The goal of preprocessing is to improve

the quality of data for the subsequent steps.

The results of classifier algorithms will be mis-

leading and adversely affected when the qual-

ity of the data is poor. Thus, preprocessing

and preparing the data is crucial in the analy-

sis of raw SELDI-MS proteomic data.

Most of the published studies have used

the software provided by the SELDI manu-

facturer for preprocessing. The SELDI soft-

ware detects the locations and intensities of

the proteins in each of the samples and carries

out important preprocessing steps like base-

line subtraction, intensity normalization, peak

alignment, and peak detection. The criteria

specified by the SELDI operator are used to

filter the peaks.

Fig. 1. Serum mass spectra. The top panel depicts the mass spectrum of a serum sample collected from a

patient with inadequate heart function (test) and the bottom panel shows that of a patient with adequate func-

tion (control). Please note that the m/z for many peaks in both spectra appear to be similar, whereas others

are different between the two patients.These spectra have been obtained using weak cation-exchange surfaces

without prefractionation of serum samples.

Baseline subtraction removes the electronic

and chemical noise. It is typically performed

in two steps. First, the baseline is estimated

using either parametric or nonparametric

methods. Then, the estimated baseline is sub-

tracted from the original mass spectrum. For

example, in the SELDI software available by

Ciphergen, baseline subtraction is performed

by applying a filter window on the mass spec-

trum, estimating the average or minimum

intensity within the window, and then moving

the window across the spectrum to estimate

the overall baseline. The size of the moving

window is a critical parameter that signifi-

cantly changes the outcome of the de-noising

step. To date, there are no studies comparing

the effectiveness of the available baseline

reduction techniques and it appears that

researchers optimize this step in a heuristic

manner that works best with their own data

sets. Although most studies are focused on

reducing the low-frequency noise in the spec-

tra, there have been many attempts to charac-

terize and subtract the high-frequency noise

components as well.

With baseline reduction completed, normal-

ization is the next step. Because a peak in a

spectrum describes only the relative amount of

a protein, normalization is done to ensure

meaningful comparisons across spectra. Once

preprocessing is performed, the peaks obtained

03_Jortani.qxd 11/21/06 3:07 PM Page 16

Volume 2, 2006________________________________________________________________ Clinical Proteomics

Data Mining In Proteomic MS ________________________________________________________________ 17

are analyzed by further dimension reduction

techniques.

Feature Selection

The Curse of Dimensionality

Most of the problems in data mining are

related to the size and the complexity of the

data sets. In data mining, a data set is referred

to as a file or a table of rows and columns,

where the rows are the records or cases and

columns are the dimensions or features or

attributes of the data. A data set with n dimen-

sions is referred to as n-dimensional. Typical

data mining classifiers are suited to problems

where the number of features is few and the

number of samples is large like the data avail-

able in retailers and manufacturers.

The straightforward approach would be

to use the normalized intensity of every

peak present in the spectrum as a feature.

Unfortunately, in proteomic applications the

number of features (dimensions) is very large

(e.g., 15,000 protein peaks) and the number

of data points is comparatively small (e.g.,

150 patients). The data samples that have

extremely high number of measurable quan-

tities are referred to as high-dimensional data.

If a one-dimensional data space has n samples,

then a k-dimensional data space must have n

different samples to achieve the same density.

However, it is often very difficult to obtain

this level of density in mass spectroscopy

applications.

This problem is known in data mining as

the curse of dimensionality. As the dimension-

ality of the input data space (i.e., the number

of features) increases, it becomes exponentially

more difficult to fit robust decision models (9).

It is seen that the data often contains noisy

features with very little or no information

value and therefore result in the development

of poor and misleading classifiers. Hence, it is

simply a practical necessity to prescreen from

among a large set of input variables those that

are of likely utility for predicting the outputs

of interest.

Prescreening the spectra to identify peaks

(or features) is known as feature extraction.

Feature extraction is typically done by bin-

ning. According to this process, each group of

m/z points that falls within a bin is described

by a value such as its average or maximum

intensity. Subsequently, the characteristics of

these bins (i.e., bin m/z location and esti-

mated intensity) are used as features for

data mining. The bins can be independent or

Fig. 2. Typical flowchart of the critical steps in data

mining and examples of techniques available for each

step. The ultimate goal is to draw conclusions based

on the data set.As shown, in general, there are several

options for each step. Sometimes the researchers may

consider using more than one particular approach in

each step.

03_Jortani.qxd 11/21/06 3:07 PM Page 17

Clinical Proteomics________________________________________________________________ Volume 2, 2006

18 _______________________________________________________________________________ Thomas et al.

overlapping, equal-size, or adaptive. By

changing the bin size and binning method,

the researcher can empirically optimize the

feature extraction process.

Selecting and using only the significant fea-

tures for the modeling process makes the

entire data mining process more accurate and

effective. The data mining algorithm would be

faster for a data set comprising of fewer and

more relevant peaks and would result in

simple and meaningful results. Thus, it is

essential to remove the irrelevant and redun-

dant features to build better models. However,

it should be noted that feature selection pro-

cess does not always guarantee successful

selection of peaks for the classification prob-

lem. Therefore, it is necessary to validate the

features that are selected by increasing the

data sample size.

Feature Selection Techniques

Advances in machine learning have led to

the development of automated feature selec-

tion tools. There are two types of feature

selection techniques in practice today. One

type analyzes each feature independently

and eliminates features one at a time based

on how that feature correlates to the target.

Independent feature selection is a simple,

straightforward, and fast process. However,

it is often the case that a group of features

together correlate more strongly with the

desired target output. Consequently, the

assumption of feature independence can be

rather limiting. To address the previously

mentioned limitation, feature selection tech-

niques have been proposed where the features

are analyzed in groups/subsets. The correlation

between the groups of features is considered

with relation to the target output. Although the

process is computationally intensive, it capi-

talizes on the important feature interrelation-

ships discovering critical information that is

typically lost with the independent feature

analysis.

Examples of independent feature selection

techniques (also called filter methods) are

receiver operating characteristics (ROC) anal-

ysis, statistical tests (i.e., t-test, Wilcoxon test,

test), wavelet transforms, information gain,

and so on. The grouped feature selection tech-

niques include stepwise feature selection,

genetic algorithms (GAs), correlation-based

feature selection, and principal component

analysis (PCA), to name a few. Feature selec-

tion is either pursued as a separate step before

decision modeling or as a step conducted in

parallel with decision modeling (called wrap-

per method) (9). Wrappers use the classifica-

tion algorithm as an integral part of the

feature selection process. In other words, the

feature selection process is optimized with

respect to the particular classifier at hand. The

following sections provide a brief description

of some feature selection techniques that have

been proposed for SELDI-MS proteomic anal-

ysis. Independent and grouped feature analysis

techniques are presented separately.

Independent Feature Selection

Statistical techniques based on t-, F, and χ

test statistics have been used in feature selec-

tion of SELDI-MS data. Liu et al. (21) used cri-

teria based on the previously listed statistics

for identifying discriminatory peaks from the

ovarian cancer data set (22). Normalization of

data during preprocessing was required prior

to performing t-test, whereas the other two

techniques were independent of this step.

Contrary, χ

test statistics were used for select-

ing the features to be used in the neural net-

work model for the detection of renal cancer

(23). Wilcoxon test was used by Sorace et al.

(24) for the ovarian cancer data set to rank fea-

tures and to understand the underlying prop-

erties of the data. Kozak et al. (25) performed

t-test and Wilcoxon test on the data set com-

prising ovarian and benign ovarian tumors.

ANOVA was used by Wagner et al. (26) on a

prostate cancer data set. The peaks prevalent

03_Jortani.qxd 11/21/06 3:07 PM Page 18

Volume 2, 2006________________________________________________________________ Clinical Proteomics

Data Mining In Proteomic MS ________________________________________________________________ 19

in less than 30 samples were removed before

applying feature selection to the remaining

peaks. Mann-Whitney test has also been used

to rank features (27).

Similarly, ROC analysis has been exploited

by Adam et al. (28) and Qu et al. (29) as the

criteria to rank the peaks for healthy and

benign cases in prostate cancer detection

before classification using a DT. In both stud-

ies, a predefined threshold of the area under

the ROC curve (AUC) was selected as the

cutoff to determine significant peaks for fur-

ther classification. Briefly, AUC is a perfor-

mance index that summarizes the ROC curve.

The area under the ROC curve is always

bounded between 0.5 and 1. An area of 0.5 for

a peak suggests that it had no discriminatory

power and 1 suggests its ability to perfectly

differentiate between two groups. In the above

studies (28,29), there were no peaks with per-

fect AUC, which implied that no peak was

independently capable of discriminating per-

fectly the groups. ROC analysis will be dis-

cussed further under performance evaluation.

Mutual information is a feature selection

technique that is based on the information

content present in the features (30). Thus far, it

has been used in analyzing proteomic mass

spectral data for the prediction of lung cancer

(31). If a particular feature is strongly corre-

lated with the target output, it should have

high mutual information. The features with

low information content are eliminated in this

method. Depending on the underlying statis-

tical distribution of the data, mutual informa-

tion captures general statistical dependencies

between each feature and the target output.

Therefore, mutual information is better suited

to nonlinear decision analysis.

Wavelet transforms, popularly used in med-

ical imaging and general image processing,

have been successfully applied by Zhu et al.

(32) to proteomic data before being classified

with a recursive tree-based algorithm. Wavelet

analysis attempts to decompose the proteomic

spectra using multiscale filter banks. After

decomposition, large wavelet coefficients were

condensed to more significant and informative

coefficients that can distinguish between the

two groups based on their within-group and

between-group sum of squares (32). Wavelet

transform has also been used by Qu et al. (33)

for feature selection of SELDI-MS data prior to

constructing a linear discriminant analysis

(LDA) function. The objective of Qu’s study

was to determine if SELDI-MS could be used

for early prostate cancer detection.

Grouped Feature Selection

Stepwise analysis is a very popular linear

feature selection algorithm based on concepts

in linear statistics. As its name suggests, fea-

tures are reviewed and selected in a step-by-

step manner and at each step the variables are

evaluated to determine which contributes most

in differentiating between the groups. Stepwise

analysis compares the means of the different

peaks and selects the peaks with the maximum

difference in the mean values in the groups.

The next peak is added such that the power of

the model increases in combination with the

first one and this process continues and the

result of the “forward stepwise analysis” is a

list of peaks in their order of significance. This

process can also be done in reverse (backward

stepwise elimination). Initially, all variables are

included in the model. The variable that least

contributes to the classification is eliminated.

Thus, only highly contributing variables are

retained in the model. Li et al. (34) used for-

ward and backward stepwise feature selection

available in the ProPeak software to rank the

peaks for the breast cancer data. Stepwise

analysis and Wilcoxon tests were repeatedly

applied on the ovarian cancer data in ref. 24 to

select the features and the classification rules

weredeveloped by visual inspection and bin-

ning based on the p-values.

The simplicity and ease in understanding

the results of stepwise analysis has made it a

03_Jortani.qxd 11/21/06 3:07 PM Page 19

Clinical Proteomics________________________________________________________________ Volume 2, 2006

20 _______________________________________________________________________________ Thomas et al.

very popular tool in feature selection. How-

ever, the stepwise method takes into consider-

ation only the linear relationships between the

peaks and does not account for any nonlinear

dependencies between them. Thus, stepwise

analysis cannot be considered an optimal

method if the features selected will be used to

develop a nonlinear model in ANN and simi-

lar nonlinear decision modeling algorithms. If

nonlinear components are present in the data

and they are important in the classification

problem, then linear feature techniques would

be less suited.

In such situations, computationally inten-

sive algorithms like those employed by GAs

would be better feature selection tools. GAs,

developed by researchers in the field of artifi-

cial intelligence (35) are based on certain

aspects of natural evolution, genetic inheri-

tance, and survival of the fittest. GAs explore

all possible subsets to obtain a set of features

that will discriminate between the groups of

training data. By operating on feature sets in

parallel, GAs obtain a global optimal result

and are known to be superior in generating

robust results for feature selection when pre-

sented with high-dimensional data. However,

the use of GAs is a computationally complex

process, especially when the number of fea-

tures is very large.

It should be noted that a GA does not

always guarantee success. Being a stochastic

system, there are situations when a fast con-

vergence occurs, which may halt the process

of evolution. There are practical limitations on

the number of iterations (generations) of a GA

and unlimited number of population size,

which may cause the GA to converge to a

local optimal solution. These algorithms are

nevertheless robust search algorithms that are

very powerful in discovering solutions for

complex high-dimensional problems where

the search space is complex and poorly under-

stood and where traditional methods fail. In

general, GAs maintain and evaluate potential

solutions, generate better and robust solutions,

and thereby improve the quality of decision

making. Conrad et al. (36) and Petricoin et al.

(37) used GAs for peak selection and self-

organizing maps for classification of SELDI-

MS proteomic data in cancer diagnosis.

PCA is an unsupervised feature selection

technique used to summarize data in high-

dimensional space into a few dimensions. Each

dimension or principal component represents

a linear combination of the original variables.

The first principal component accounts for

most of the variability of the data. The next

principal component accounts for the variabil-

ity not accounted for by the first component

and so on. Typically, the first few principal

components are identified and then the data

set is projected onto these components for

dimension reduction. There are several appli-

cations of PCA for feature selection in MS data

(38–41). For example, Lilien et al. used a spe-

cific algorithm called Q5 in which PCA was

used for feature selection and LDA for model

development (38).

Other than an exhaustive search of every

possible feature combination, there is no par-

ticular feature selection strategy that guaran-

tees optimal results. Consequently, it is

advised to explore a variety of strategies given

a particular proteomic data set. For example,

Liu et al. (21) investigated several feature

selection and classification techniques using a

publicly available ovarian cancer data set.

Specifically, three feature selection techniques

(χ

method, t-test, and correlation analysis) were

used in their study. The features selected by

these techniques were then fed into diverse

classification methods such as support vector

machines (SVM), DTs, k-nearest neighbor

(KNN), and naïve-Bayesian algorithms. Results

with and without the feature selection phase

were acquired to determine the importance

of the feature selection process. The study

clearly demonstrated that feature selection

is a critical step but it should always be

03_Jortani.qxd 11/21/06 3:07 PM Page 20

Volume 2, 2006________________________________________________________________ Clinical Proteomics

Data Mining In Proteomic MS ________________________________________________________________ 21

considered in relation to the classification

step. For this particular study, SVM classifi-

cation algorithm and KNN reported the best

classification accuracy (100%) when used

with the peaks selected by correlation-based

feature analysis. In contrast, models built by

randomly selecting features from the data

resulted in poor accuracy.

In another study, Li et al. (42) compared the

performance of a SVM classifier using the filter

approach (a statistical test using a distance

measure) and GAs in ovarian cancer detection

and diagnosis. The main goal of the study was

to demonstrate the feasibility of applying arti-

ficial intelligence techniques on serum pro-

teomic data. In this study, GAs emerged as the

superior feature selection strategy when used

in combination with a SVM classifier.

To summarize, there is no general guide-

line on which feature selection strategy

should be used. It is generally understood

that grouped feature selection techniques

that are capable of capturing nonlinear rela-

tionships among the available features

should be used in combination with non-

linear models. Typically, researchers should

investigate various techniques to empirically

optimize the feature selection process. Once

the features are selected, data modeling is the

next step.

Classifiers and Data Modeling

Humans and animals acquire the ability to

learn and recognize through interaction with

the environment. Learning from data has

been an area of interest for researchers in

statistics and computer science. Machine

learning algorithms are able to make infer-

ences about a sample of data through famil-

iarization and repeated interaction with the

data. These algorithms vary in their training

techniques, final goal, and representation of

data.

A learning process would typically com-

prise of the task of learning and developing

rules or functions from the given data set of

samples. The development of mathematically

accurate rules and functions to describe data

is called data modeling. The model developed

identifies the properties of the different classes

and what separates them to make a correct

classification. In the next phase, called testing,

the developed model is validated with new

data to verify that the model produces accu-

rate results. The learning phase and estimation

of the model is implemented and described

using different learning methods or algo-

rithms.

Afunction that describes or approximates

the data is of the form f (X, w) where X is the

input and w is a parameter of the function.

The function f can be linear or nonlinear.

Machine learning classifiers that are based on

nonlinear decision function include ANN, self-

organizing maps (SOMs), and GAs. LDA is an

example of a linear classifier.

Machine learning algorithms are of two

types—supervised and unsupervised. In super-

vised learning (also called “learning with a

teacher”), prior knowledge is available about

the class to which each case (sample) belongs.

The training data set comprises of input

values and their corresponding output classes

(provided by teacher). During the training

phase, the training data is used to determine

how the features are to be selected, weighed,

and combined so as to discriminate between

the classes. The testing phase involves appli-

cation of the weighted features to classify a

new test data whose class is not known and

which the decision model has not seen before.

Thus, the goal of classification methods is to

build models of the data set in hand and use

that model to classify new samples. The pro-

cess of learning would involve creating a

model so that the predictions of the model are

close to the desired target. If the model is able

to classify new data correctly we have reason

to believe that it is a good model. A wide

range of algorithms has been developed for

03_Jortani.qxd 11/21/06 3:07 PM Page 21

Clinical Proteomics________________________________________________________________ Volume 2, 2006

22 _______________________________________________________________________________ Thomas et al.

supervised learning (DTs, SVM, logistic regres-

sion, et al.).

In unsupervised learning (“learning with-

out a teacher”), the group to which each

sample belongs is unknown or ignored and

data is grouped together based on similarity

measures. The learning process involves no

teacher and the algorithm must identify pat-

terns in the data. Often, unsupervised learning

may lead to more than one possible solution.

Clustering and Kohonen’s SOMs are typical

examples of unsupervised learning that have

been used in studies analyzing mass spec-

troscopy data. ANNs, on the other hand,

come as supervised and unsupervised learn-

ing algorithms.

In both learning techniques, the goal is to

predict (classify) or describe the data by devel-

oping models of the data, which are then used

to classify or describe new cases. If the data

has just two or three features, then classifying

data would be easy. However, developing

models can be a daunting task if there are

many features to analyze. High-dimensional

data is not only hard to visualize, but all pos-

sible combinations should be considered by

exhaustive searching techniques during the

training phase when the model is developed.

A large number of dimensions with very few

samples leads to what is often referred to as

over-fit or over-trained models. Over-fit models

cannot generalize and fail to classify new

cases with the desired accuracy.

Supervised Classifiers

LDA is a supervised linear modeling

approach and was one of the first to be

applied to proteomic mass spectral data. LDA

is a noniterative and deterministic classifica-

tion approach. LDA analysis computes exact

solutions, it is simple to implement, and easy

to understand. LDA is typically inefficient in

classifying high-dimensional data and thus

feature selection techniques should be applied

before any modeling can be done. LDA

captures potential linear relationships in the

data and represents it in the form of a linear

function. However, nonlinear and complex

relationships cannot be fully captured when a

linear classifier is used.

LDA has been used in SELDI-MS pro-

teomic studies conducted by Lilien et al.

where the LDA was a part of their Q5 algo-

rithm (38). Wagner et al. (26) analyzed prostate

cancer data using both LDA and QDA

(quadratic discriminant analysis). The feature

reduction step helped in reducing the number

of peaks by 97%. In another study, Qu et al.

(34) analyzed prostate cancer data comprising

45,538 peaks. They developed an LDA model

using 12 peaks selected by wavelet transfor-

mation and a procedure based on Mahanobis

distance.

DT analysis is another popular supervised

classification technique (9). The resulting

model is in the form of a hierarchical tree

structure. The main advantage of DTs is that

they are expressed as a set of rules. As they

are visually presented in an easy to interpret

form, DTs are very popular in several domains

including proteomics. A DT’s goal is to iden-

tify a set of variables (peaks) that can be used

to classify cases or samples into specific

groups. A DT performs classification through

a systematic process referred to as recursive

partitioning (9). At each node of the tree, a

test (presence/absence/intensity of peaks) is

applied to one or more variables (peaks) that

will have one of two outcomes. The outcome

will lead to a split into a leaf node or to

another decision node where another test is

applied. Peaks are selected for a split based on

a cost function, which is the measure of het-

erogeneity of the descendant nodes. The cost

function is often an entropy measure or gain

in information. The splitting is made and the

tree model is built until there is no gain in

information associated with a split or if the

there are no more nodes left. Finally, classifi-

cation of new test data is done by following

03_Jortani.qxd 11/21/06 3:07 PM Page 22

Volume 2, 2006________________________________________________________________ Clinical Proteomics

Data Mining In Proteomic MS ________________________________________________________________ 23

the tree rules (i.e., by moving through the tree

until a leaf is encountered).

Popular programs for constructing DTs are

C4.5 (43) and classification and regression tree

(CART) (44). The C4.5 algorithm was used by

Won et al. (37) to classify the data set of 36

samples and 119 peaks into renal cell carci-

noma (RCC) and healthy patients. The 119

peaks were identified by preprocessing using

the Ciphergen Protein Chip software. The DT

identified five discriminatory peaks for model

development. A CART model was developed

by Markey et al. (46) to classify 41 serum

samples with 21 peaks into lung cancer and

controls. The peaks used in this model were

identified by mass spectrometry. Adam et al.

(28) used CART to classify serum proteomic

patterns identified by Ciphergen System soft-

ware (i.e., the Biomarker Wizard), into prostate

cancer, benign prostate hyperplasia (BPH),

and controls. Three hundred and twenty-six

serum samples from 167 PCA patients, 77 BPH

patients, and 82 healthy men were used in this

particular study. More than 60,000 peaks in the

range of 2000–40,000 Da were selected by the

SELDI software and was reduced to 772 by

preprocessing and then to 124 by peak selec-

tion using ROC analysis. Finally, the DT

selected nine peaks to develop the model.

Biomarker pattern software (BPS) based on the

CART decision model was used by Zhang

et al. (47) to classify 156 urine samples into

transitional cell carcinoma, benign urogenital

diseases, and controls. The process started with

peak detection phase using the Biomarker

Wizard software. A signal-to-noise ratio of 5

was used to filter the peaks that were in the

range 2000 to 50,000 Da. Four hundred peaks

produced by peak detection were used in DT

construction. The training algorithm identi-

fied five discriminatory peaks and developed

the tree model. BPS was also used by Kang

(48) to identify serum biomarkers that distin-

guish between severe acute respiratory syn-

drome (SARS) and non-SARS samples.

Overall, DTs are easy to interpret and can be

represented in the form of if–then rules. The

training and testing times are reasonable when

compared with neural networks and they can

handle large number of features. Despite these

advantages, they are not recommended when

there are large amounts of missing data and in

problems involving complex data distributions.

The instability associated with DTs causes small

changes in data to result in entirely different

decision rules. Therefore, interpreting DT results

should be exercised with caution.

Thus far, the discussion was on supervised

linear classifiers. Another popular classifier

that has been used in SELDI-MS protein data

analysis is the ANN (49). ANNs are powerful

nonlinear classifiers, based on artificial intelli-

gence principles. Inspired by the human brain,

they comprise of highly connected and non-

linear units referred to as neurons that are

connected by weights. The interunit connec-

tion weights stores the processing ability of

the neural network. This is obtained by the

process of learning from the training data

set. There are two different types of neural

networks—supervised neural networks and

unsupervised neural networks.

By far, the most popular ANN architecture

in medical decision making has been the tradi-

tional backpropagation neural network, intro-

duced by Rumelhart et al. (50). A simple

representation of a neural network is as a set

of interconnected layer of neurons with

weighted interconnections reflecting various

influences of each neuron on the others (Fig. 3

illustrates a simplified neural network model).

Training is performed by using a data set with

known inputs and outputs. In the backpropa-

gation training algorithm, the neural network

output is compared with its expected output

for each training example. Computation starts

at the output layer and propagates backward

to the hidden layer(s). The difference between

the desired output and actual output (error) is

calculated and propagated back through the

03_Jortani.qxd 11/21/06 3:07 PM Page 23

Clinical Proteomics________________________________________________________________ Volume 2, 2006

24 _______________________________________________________________________________ Thomas et al.

network. The neural network weights are

adjusted in an iterative manner until the net-

work achieves a predefined threshold of min-

imum squared error for the whole training set.

As with all supervised classifiers, develop-

ment of an ANN involves both the training

and validation phases of the network. During

the learning phase, the data is presented

repeatedly to the ANN and it learns from that

by storing the decision rules as weights. Fol-

lowing the training phase is the validation

phase where new cases are presented for clas-

sification. Rogers et al. (23) used 48 data sam-

ples from patients with RCC, 38 controls, and

20 benign cases to train a feed-forward neural

network model. They reported sensitivity and

specificity values in the range 98.3 to 100%

during training and 81.8 to 83.3% during test-

ing. However, further testing was performed

later to assess the robustness of the model

with a group of 80 samples to get sensitivity

and specificity in the 41 to 76.6% range, which

called for further evaluation of contributing

factors. SELDI-MS coupled with ANN was

used by Mian et al. (51) to identify protein pat-

terns associated with either control/drug treat-

ment for chemotherapy-sensitive cells or

response/nonresponse for chemotherapy-

resistant breast cancer cells, respectively for two

popular chemotherapeutic drugs. The model

developed was able to classify the test data as

resistant or as sensitive to a particular drug

and discriminate between chemotherapy-

sensitive and -resistant breast cancer cells with

high accuracy. Ball et al. (52) used multilay-

ered perception ANN with a backpropagation

algorithm to mine the SELDI-analyzed data of

tumors and controls. Poon et al. (53) used

neural networks to discriminate hepatocellular

carcinoma (HCC) from chronic liver disease

(CLD). Two hundred and fifty significant

peaks identified by significance analysis of

microarrays software were used to develop a

feed-forward backpropagation ANN model.

ROC curve analysis showed that the ANN

was useful in differentiating HCC and CLD

cases regardless of serum concentrations. They

reported a sensitivity and specificity for HCC

and CLD cases as approx 95% and approx

90%, respectively.

However, ANNs are not devoid of limita-

tions. The main criticism regarding ANNs is

their “black box” nature. It is very difficult to

determine how the neural network makes the

decisions in classifying data. An additional

disadvantage is their low computational effi-

ciency during the training phase. The optimal

theoretical size of data required for training

and ANN is almost never met in medical

research. Small sample size often leads to over-

fitting. Furthermore, the architecture of the

ANN for getting the best results is subjective.

Changes in the ANN architecture, training

parameters, and starting point of training will

result in different models and solutions. The

training algorithms often get trapped in the

local minima resulting in suboptimal results.

Despite all the issues surrounding the archi-

tecture and training of ANNs, this group of

classifiers is known to produce generalized

results and often report better accuracy than

traditional statistical techniques in medical

diagnosis. In addition, neural networks can

handle problems with large amounts of

diverse features and are known to perform

well with complex data distributions.

Fig. 3. A schematic representation of an artificial

neural network.

03_Jortani.qxd 11/21/06 3:07 PM Page 24

Volume 2, 2006________________________________________________________________ Clinical Proteomics

Data Mining In Proteomic MS ________________________________________________________________ 25

Unsupervised Classifiers

Clustering is an unsupervised learning

technique (9), wherein data samples are

grouped into clusters, based on a measure of

association so that similar groups are in one

cluster. Input to cluster analysis is a set of

samples and a measure of association (simi-

larity or dissimilarity between two samples).

The output is a group of clusters and a gen-

eralized description of each cluster. Hierar-

chical clustering is a popular clustering

algorithm (9) that has been used in several

SELDI studies. The data is grouped into a

sequence of clusters in a bottom-up manner

and the result is displayed in the form of a

dendrogram or a tree structure. The number

of clusters in the data is not prespecified in

the problem. Agglomerative hierarchical clus-

tering is a type of hierarchical clustering

algorithm where each sample is regarded as a

cluster. In the agglomerative clustering tech-

nique, pairs of clusters closer to each other are

merged based on their similarity measure and

this is repeated until the entire data is in one

cluster.

Poon et al. (53) used two-way hierarchical

clustering to differentiate HCC from CLD.

The result of the study reported was that the

algorithm separated HCC and CLD cases into

two major clusters. However, there was no

mention of the model evaluation. Purohit et

al. (38) performed hierarchical cluster analy-

sis for classifying data into diseased and

healthy. The PCA feature selection method

was first applied on the data. This combina-

tion of PCA and cluster analysis could cor-

rectly identify 68% of the patients and 100%

of the healthy group. They indicated the pos-

sibility of the diseased people to be in a dif-

ferent disease stage.

SOM is a nonlinear clustering technique and

was first introduced by Kohonen (54). Similar

in design to the human brain, an SOM consists

of interconnected neurons and has two layers;

the input layer, which contains the raw data

and the output layer, which comprises of the

clusters. The process of learning involves suc-

cessive passes through the input layer until

no new information is obtained on subse-

quent passes. The result of this process is a set

of clusters that best represent the differences

in the data. SOMs follow unsupervised learn-

ing and depend on the self-organizing nature

of the data for the formation of best set of

clusters.

SOMs have been used in the Petricoin

et al. (22) study in combination with GAs to

identify ovarian cancer biomarkers and to

classify the samples into healthy, benign, and

cancer cases. The clustering algorithm was

trained on 50 healthy women and 50 women

in different stages of ovarian cancer and tested

on 116 new cases, yielding a sensitivity of

100% and specificity of 95%.

In samples with categorical data, KNN may

be used to compute the similarity between

samples and clusters. This technique is based

on the distances from immediate neighbors.

The class of an input pattern is chosen as the

class of the majority of its KNN. Euclidean

and Mahanobis distances are often used as the

distance metrics (9).

There are several clustering algorithms

available in the literature and could easily con-

found a new researcher trying to get a suitable

algorithm. All clustering algorithms will pro-

duce clusters in a given data irrespective of

whether or not they exist. Therefore, cluster

algorithms should only be applied on a data

set that is expected to contain clusters. Once

clusters are formed out of the data, they

should be evaluated using existing validation

techniques (external and internal criterion

analysis) for cluster validation. However,

existing validation techniques are subjective

in nature and data analysts should keep in

mind that there is no best clustering algo-

rithm. Therefore, empirical study is advised

by exploring several algorithms on a given

data set.

03_Jortani.qxd 11/21/06 3:07 PM Page 25

Clinical Proteomics________________________________________________________________ Volume 2, 2006

26 _______________________________________________________________________________ Thomas et al.

Combining Classifiers

Although most of the recent proteomic

studies have been performed using individual

classifiers, the approach of combining classifiers

is a promising strategy to reduce the classifica-

tion prediction error and to improve decision-

making performance. The final classifier is a

set of base classifiers, each of which makes its

own classification and they are combined to

constitute the single classification result of the

entire classifier.

Breiman (55) refers to classifiers as unsta-

ble if a small change in data causes large

changes in classification and they often have

high bias and low variance. The opposite is

true for stable classifiers. DTs and neural nets

are considered to be unstable classifiers,

whereas LDA and KNN are considered to be

stable. If classifiers are combined, the variance

and error are usually lowered. Bagging and

boosting are two such techniques that have

been used for combining classifiers in data

mining and they have also been applied in

proteomic data analysis.

Boosting of weak classifiers is performed by

taking a weighted vote of each classifier.

Boosting applies the training algorithm

sequentially to the training sample and the

sample is reweighed to give importance to

cases that are not correctly predicted. A

weighted majority vote of the classifiers is

taken for the decision making. This method

often results in a stronger classifier and is very

effective in DTs, which are considered to be

unstable classifiers, as it boosts the perfor-

mance of existing weak learners, reduces the

test error and variance, avoids over-fitting,

and thus leads to an improvement of perfor-

mance. For example, a boosted DT was used

in the early detection of prostate cancer in the

studies conducted by Qu et al. (29). The study

aimed at developing a classifier to discrimi-

nate men with prostate cancer from those with

BPH and controls.

Bagging is another technique that has been

used to combine predictors and was first used

by Breiman in data mining applications. Many

replicates of the available data set having the

same size as the original data set are drawn

with replacement from the training sample

(called bootstrap samples). The learning algo-

rithm is then applied to each bootstrap sample

and validation is then performed on the sam-

ples where training is not done. The final clas-

sification is completed by taking a majority

vote where each classifier has equal weight in

voting. When the individual results from each

classifier vary considerably from one another,

taking an average of the results will result in a

more accurate prediction. However, if the

results are similar, averaging the results may

not result in a better performance. Thus, the

advantage of bagging is clearly when the per-

formance of individual classifiers is uncorre-

lated. Izmirlan (56) used bagging in a proteomic

random forest classification algorithm where

DTs were bagged using 632 cross-validation

and randomly selected feature sets.

The result of combining classifiers often

improves performance, especially when com-

bining classifiers that do not think alike and

have low correlation. Using a combined

approach, individual classifiers complement

each other by capturing information that indi-

vidually they miss.

Sampling

One of the major challenges faced in the

application of machine learning algorithms to

medical data is the validation of a trained

model with new test data. The process of deci-

sion modeling requires that the model be

developed by training on a certain set of the

data (training set), which is followed by the

validation of the model on another set of data

not previously seen during training (testing

set). The obvious way to handle this issue is by

dividing the data into train and test sets before

model building by stratified random sampling.

03_Jortani.qxd 11/21/06 3:07 PM Page 26

Volume 2, 2006________________________________________________________________ Clinical Proteomics

Data Mining In Proteomic MS ________________________________________________________________ 27

Stratified random sampling was used in refs. 28

and 45. These studies report the performance

of the decision models based on a single train-

test data split.

However, medical data is quite often very

difficult and expensive to acquire. Thus, there

are not enough available cases to divide into

train-test subsets. Furthermore, the noise

inherent in most medical data and the com-

plex relationships between features requires a

sample size large enough to efficiently model

the data with accuracy. In addition, the size of

the test set controls the statistical power and

confidence in the developed decision model.

Consequently, sophisticated sampling strate-

gies are required to capitalize on the available

data (57). The following is a brief description

of data sampling strategies recommended in

medical decision making.

Cross-Validation

Cross-validation is by far the most popular

data sampling strategy. The data is randomly

divided into two sets. The decision model is

trained on the first and tested on the second.

This random splitting process is repeated sev-

eral times to reduce the selection bias. The

average of all the test estimates gives the

average error of the model. If the data set

used for training is too small, the model may

not be able to predict test cases well. A small

test set may not result in an accurately vali-

dated classifier and can have a large error

rate. Consequently, different train–test ratios

(e.g., 50–50%, 75–25%, etc.) are explored with

cross-validation.

A common implementation is the k-fold

cross-validation. The data is partitioned into

k-disjoint sets. Training of the classifier is done

on the k-1 sets and testing on remaining one

data set. This is done for all the k-subsets

resulting in k-models and the estimated error

will be the average of the k-error rates. For

example, 10-fold cross-validation divides the

data into 10 groups. Nine groups are used for

training and testing is done on the left out-

group. This is repeated 10 times until each of

the 10 groups has served as the test group.

The average test error of the 10 groups is the

final test error estimate and gives an approxi-

mate idea of the quality of the model for the

classification of the data. K-fold cross-validation

requires the careful random stratification of the

10 groups. Ten-fold cross-validation was used

in ref. 57 to discriminate between early-

stage melanoma patients with and without

melanoma recurrence. The sensitivity and

specificity were 72 and 75%, respectively.

Wagner et al. (26) stressed the significance of

cross-validation with various decision models.

Their study showed results based on a 90- to

10%-data split, randomly repeated 100 times.

Zhang et al. (47) employed the same data sam-

pling strategy.

A noteworthy special case is the popular

leave-one-out cross-validation approach, a spe-

cial case of the k-fold cross-validation imple-

mentation. If the data comprises of k-samples,

the data model is trained using the k-1 samples

and tested on the remaining one sample. This is

repeated k times until every sample has served

as a test case. The average error on the k sam-

ples is the estimated test error. The main advan-

tage of the technique is that the decision model

uses almost all available cases for training with-

out compromising the statistical significance of

the testing phase. However, the technique can

be time consuming with large sample sizes and

elaborate decision models. Therefore, it is used

predominantly in resampling of small data sets.

There are several published examples of

studies using variations of the cross-validation

sampling scheme (see refs. 32,38,39,42,46,59).

In a comparative study, Zhu et al. (32) used

both leave-one-out and k-fold cross-validation

to assess the validity of their DT model and

achieved an error rate of 14.5 and 20% for

leave-one-out and k-fold cross-validation,

respectively showing that leave-one-out has a

lower bias than the latter.

03_Jortani.qxd 11/21/06 3:07 PM Page 27

Clinical Proteomics________________________________________________________________ Volume 2, 2006

28 _______________________________________________________________________________ Thomas et al.

Bootstrap

The bootstrap method involves the genera-

tion of a training set by sampling with replace-

ment. This is done n times for a data set of n

cases. The data is trained on the bootstrap set

and tested on the original set. It is recom-

mended that the process is repeated many

times (>200). The final performance estimate

is the average of all bootstrap estimates. This

is a computationally intensive process even for

small size data sets. Mian et al. (51) used boot-

strap validation with 86 independent ANN

models. The variation in the model’s perfor-

mance, accuracy of classifying test data, and

identification of potential outliers were

assessed through the bootstrap resampling.

Performance Assessment of Models

The last phase of the data mining process is

the assessment of the models developed by

the previously described machine learning

algorithms. The different methods for evaluat-

ing the models’ ability to classify new test

cases are discussed below.

Accuracy

Accuracy of classification is calculated by

taking the ratio of the number of correctly clas-

sified samples to the total number of samples

in the test data. However, when the prevalence

of a particular class is higher than the other

class, the majority class will cause a bias in the

result. In such a scenario, the accuracy measure

can be misleading. SELDI-MS-based proteomic

studies that have used accuracy to report results

include those in refs. 25,31,39,42,51, and 52.

Sensitivity/Specificity

In two-class samples, there are four possi-

ble outcomes when the decision model is

tested. They are true-positive, true-negative,

false-positive, and false-negative results. Sen-

sitivity (true-positive rate) is the ratio of the

number of correctly classified positive samples

over the total number of positive samples.

High sensitivity is much desired in medical

diagnosis where the impact of wrongly pre-

dicting a diseased person as healthy is high.

False-positive rate is the probability that a

healthy subject is wrongly classified as dis-

eased (referred to as specificity). High speci-

ficity is desirable where a false alarm would

result in unwanted elaborate tests and treat-

ments. Ideally, for perfect classification, both

sensitivity and specificity should be 1 (100%).

The clinically acceptable sensitivity and speci-

ficity depends on the application.

Several studies have reported their results

using sensitivity and specificity as the perfor-

mance indices (see refs. 23,24,28,33,45–47). The

main limitation of using sensitivity and speci-

ficity as the only evaluation indices is their

dependence on class prevalence and the deci-

sion threshold. Therefore, it is difficult to

directly compare the results of studies that are

reported using only sensitivity and specificity

measures.

ROC Analysis

Based on the classical signal detection

theory, ROC analysis reports the sensitivity

and specificity of a decision model for all pos-

sible decision thresholds. Conventionally, a

ROC curve plots the true-positive fraction (or

sensitivity) vs the false-positive fraction (or [1-

specificity]) for a wide and continuous range

of decision thresholds and provides a more

meaningful and valid measure of classification

performance. Furthermore, ROC curves can be

used to determine optimum decision levels

that maximize accuracy, average benefit, or

other measures of clinical efficacy. The AUC

gives a complete picture of the performance of

a model and can be used to compare the per-

formance of multiple models. The more the

curve is shifted to the upper left corner of the

graph (specificity = sensitivity = 1), the better the

diagnostic performance. The area index varies

between 0.5 (representing chance behavior) and

03_Jortani.qxd 11/21/06 3:07 PM Page 28

Volume 2, 2006________________________________________________________________ Clinical Proteomics

Data Mining In Proteomic MS ________________________________________________________________ 29

1.0 (representing perfect performance). Gener-

ally, a higher value of the area index indicates

better performance. More importantly, ROC

analysis is threshold and prevalence indepen-

dent. Figure 4 shows typical ROC curves.

The usage of ROC in data mining is still

below its full potential. It is highly recom-

mended that researchers embrace this tech-

nique, since comparison of results across

different studies is straightforward. ROC anal-

ysis has been utilized in several SELDI pro-

teomic studies (25,28,34,42,46).

Conclusions

Data mining is a data-driven process where

the results obtained largely depend on the

data being analyzed. The methods employed

for feature selection, classification, data sam-

pling, and performance evaluation drive the

process and alter final results. Thus, it is

recommended to explore more than one tech-

nique to make comparisons and better under-

stand the problem in hand. The promise and

opportunities of combination of data mining

approaches and the generated proteomic mass

spectrometry data for discovery of novel

biomarkers with diagnostic value is obvious.

However, caution should be exercised in

application applying various data mining

techniques in this regard. This review made an

effort to reinstate the critical issues and limita-

tion of data mining applications to be

addressed by researchers analyzing SELDI-MS

proteomic data for extracting clinically useful

information.

Acknowledgments

Funding support for AT was in part by a

grant from NIH-NCRR 5 P20 16480 (principal

investigator: Nigel Cooper).

References

1. Wilkins, M. R., Sanchez, J. C., Gooley, A. A., et al.

(1996) Progress with genome projects: why all

proteins expressed by a genome should be

identified and how to do it. Biotechnol. Genet.

Eng. Rev. 13, 19–50.

2. Coombes, K. (2005) Analysis of mass spec-

trometry profiles of the serum proteome. Clin.

Chem. 51, 1–2.

3. Rodland, K. D. (2004) Proteomics and cancer

diagnosis. Clin. Bioch. 37, 579–583.

4. Liotta, L. A., Ardekani, A. M., Hitt, B. H., et al.

(2003) General keynote: proteomic patterns in

sera serve as biomarkers of ovarian cancer.

Gynecol. Oncol. 88, S25–S28.

5. Conrads, T. P., Fusaro, V. A., Ross, S., et al.

(2004) High-resolution serum proteomic fea-

tures for ovarian cancer detection. Endocr.

Relat. Cancer 11, 163–178.

6. Yip, T. C., Chan, J. W., Cho, W. C., et al. (2005)

Protein chip array profiling analysis in patients

with severe acute respiratory syndrome identi-

fied serum amyloid: a protein as a biomarker

potentially useful in monitoring the extent of

pneumonia. Clin. Chem. 51, 47–55.

7. Coombes, K. R., Koomen, J. M., Baggerly, K. A.,

Morris, J. S., and Kobayashi, R. (2005) Under-

standing the characteristics of mass spectrom-

etry data through the use of simulation.

Cancer Informatics 1, 41–52.

8. Hong, H., Dragan, Y., Epstein, J., et al.

(2005) Quality control and quality assess-

ment of data from surface-enhanced laser

desorption/ionization (SELDI) time-of

flight (TOF) mass spectrometry (MS). BMC

Bioinformatics 15, S5.

Fig. 4. A typical ROC curve.The closer the line is to

the upper left had corner of the graph, the perfor-

mance of the test is considered to be best with the

least false-positives or -negatives.

03_Jortani.qxd 11/21/06 3:07 PM Page 29

Clinical Proteomics________________________________________________________________ Volume 2, 2006

30 _______________________________________________________________________________ Thomas et al.

9. Kantardzic, M. (2002) Data Mining: Concepts,

Methods, Models, and Algorithms. Wiley and

IEEE Press, New York.

10. Wilson, R. L. and Sharda, R. (1994) Bankruptcy

prediction using neural networks. Decision

Support Systems 11, 545–557.

11. Barr, D. S. and Mani, G. (1994) Using Neural

Nets to manage investments. AI Expert 1994;

16–21.

12. Sung, T. K., Chang, N., and Lee, G. (1999)

Dynamics of modeling in data mining: inter-

pretive approach to bankruptcy prediction.

J. Manag. Info. Sys. 1, 63–85.

13. Shaw, M. J., Subramaniam, G., Tan, G. W., and

Welge, M. E. (2001) Knowledge management

and data mining for marketing. Dec. Supp. Sys.

31, 127–137.

14. Daskalaki, S., Kopanas, I., Goudara, M., and

Avouris, N. (2003) Data mining for decision

support on customer insolvency in telecom-

munications business. Eur. J. Oper. Res. 145,

239–255.

15. Haa, S. H., Baeb, S. M., and Parkb, S. C. (2002)

Customer’s time-variant purchase behavior and

corresponding marketing strategies: an online

retailer’s case. Comp. and Indus. Eng. 43, 801–820.

16. Caskey, K. R. (2001) A manufacturing problem

solving environment combining evaluation,

search, and generalisation methods. Computers

in Industry 44, 175–187.

17. Kusiak, A., Dixon, B., and Shaha, S. (2005) Pre-

dicting survival time for kidney dialysis

patients: a data mining approach. Comp. Biol.

Med. 35, 311–327.

18. Chen, W. H., Hsu, S. H., and Shen, H. P. (2005)

Application of SVM and ANN for intrusion

detection. Comp. and Oper. Res. 32, 2617–2634.

19. Seifert, J. W. (2004) Data mining and the

search for security: challenges connecting the

dots and databases. Government Information

Quarterly 21, 461–480.

20. Barrera, J., Cesar, R. M., Ferreira, J. E., and

Gubitoso, M. D. (2004) An environment for

knowledge discovery in biology. Comp. Biol.

Med. 34, 427–447.

21. Liu, H., Li, J., and Wong, L. (2002) A compar-

ative study on feature selection and classifica-

tion methods using gene expression profiles.

Genome Informatics 13, 51–60.

22. Petricoin, M. F., Ardekani, A. M., Hitt, B. A.,

et al. (2002) Use of proteomic patterns in

serum to identify ovarian cancer. Lancet 359,

572–577.

23. Rogers, M. A., Clarke, P., Noble, J., et al. (2003)

Proteomic profiling of urinary proteins in

renal cancer by surface enhanced laser des-

orption ionization and neural-network analy-

sis: identification of key issues affecting

potential clinical utility. Cancer Res. 63,

6971–6983.

24. Sorace, J. M. and Zhan, M. (2003) A data

review and reassessment of ovarian cancer

serum proteomic profiling. BMC Bioinformatics

4, 24.

25. Kozak, K. R., Amneus, M. W., Pusey, S. M.,

et al. (2003) Identification of biomarkers for

ovarian cancer using strong anion-exchange

proteinchips: potential use in diagnosis and

prognosis. PNAS 100, 14,666–14,671.

26. Wagner, M., Naik, D. N., Pothn, A., et al.

(2004) Computational protein biomarker pre-

diction: a case study for prostate cancer. BMC

Bioinformatics 5, 26.

27. Zhukov, T. A., Johnson, R. A., Cantor, A. B.,

Clark, R. A., and Tockman, M. S. (2003) Dis-

covery of distinct protein profiles specific for

lung tumors and pre-malignat lung lesions by

SELDI mass spectrometry. Lung Cancer 40,

267–279.

28. Adam, B. L., Qu, Y., Davis, J. W., et al. (2002)

Serum protein finger printing coupled with a

pattern-matching algorithm distinguishes

prostate cancer from benign hyperplasia and

healthy men. Cancer Research 62, 3609–3614.

29. Qu, Y., Adam, B. L., Yasui, Y., et al. (2002)

Boosted decision tree analysis of surface

enhanced laser desorption/ionization mass

spectral serum profiles discriminates prostate

cancer from noncancer patients. Clin. Chem.

48, 1835–1843.

30. Tourassi, G. D., Frederick, E. D., Markey, M.

M., and Floyd, C. E. (2001) Application of the

mutual information criterion for feature selec-

tion in computer-aided diagnosis. Med. Phys.

28, 2394–2402.

31. Hilario, M., Kalousis, A., Müller, M.,

and Pellegrini, C. (2003) Machine learning

approaches to lung cancer: prediction from

mass spectra. Proteomics 3, 1716–1719.

32. Zhu, H., Yu, C. Y., and Zhang, H. (2003) Tree

based disease classification using protein data.

Proteomics 3, 1673–1677.

03_Jortani.qxd 11/21/06 3:07 PM Page 30

Volume 2, 2006________________________________________________________________ Clinical Proteomics

Data Mining In Proteomic MS ________________________________________________________________ 31

33. Qu, Y., Adam, B. L., Thornquist, M., Potter, J.

D., Thompson, M. L., and Yasui, Y. (2003) Data

reduction using a discrete wavelet transform

in discriminant analysis of very high dimen-

sionality data. Biometrics 59, 143–151.

34. Li, J., Zhang, Z., Rosenzweig, J., Wang, Y. A.,

and Chan, D. W. (2002) Proteomics and bioin-

formatics approaches for identification of

serum biomarkers to detect breast cancer. Clin.

Chem. 47, 1296–1304.

35. Holland, J. H. (1994) Adaptation in Natural and

Artificial Systems: An Introductory Analysis

With Applications of Biology, Control and Artifi-

cial Intelligence, 3rd ed. MIT Press, Cambridge,

MA.

36. Conrads, T. P., Zhou, M., Petricoin, E. F.,

Liotta, L., and Veenstra, T. D. (2003) Cancer

diagnosis using proteomic patterns. Expert

Rev. Mol. Diagn. 3, 411–420.

37. Petricoin, E. F. and Liotta, L. A. (2004) SELDI-

TOF based proteomic pattern diagnostics for

early detection of cancer. Curr. Opin. Biotech.

15, 24–30.

38. Lilien, R. H., Farid, H., and Donald, B. R.

(2003) Probabilisitic disease classification of

expression—dependent proteomic data from

mass spectrometry of human serum. J. Comp.

Biol. 10, 925–946.

39. Purohit, P. V. and Rocke, D. M. (2003) Dis-

criminant models for high-throughput pro-

teomics mass spectrometer data. Proteomics 3,

1699–1703.

40. Slotta, D. J., Heath, L. S., Ramakrishnan, N.,

Helm, R., and Potts, M. (2003) Clustering mass

spectrometry data using order statistics. Pro-

teomics 3, 1687–1691.

41. Coombes, K. R., Fritsche, H. A., Clarke, C., et al.

(2003) Quality control and peak finding from

nipple aspirate fluid by surface enhanced laser

desorption and ionization. Clin. Chem. 49,

1615–1623.

42. Li, L., Tang, H., Wu, Z., et al. (2004) Data

mining techniques for cancer detection using

serum proteomic profiling. Artif. Intel. Med. 32,

71–83.

43. Quinlan, J. R. (1986) Introduction of decision

trees. Machine Learning 1, 81–106.

44. Breiman, L., Friedman, J. H., Olshen, R. A.,

and Stone, C. J. (1984) Classification and

Regression Trees. Wadsworth International

Group, Belmont, CA.

45. Won, Y., Song, H. J., Kang, T. W., Kim, J. J.,

Han, B. D., and Lee, S. W. (2003) Pattern anal-

ysis of serum proteome distinguished renal

cell carcinoma from other urologic diseases

and healthy persons. Proteomics 3, 2310–2316.

46. Markey, M. K., Tourassi, G. D., and Floyd, C.

E., Jr. Decision Tree classification of proteins

identified by mass spectrometry of blood sam-

ples from people with and without lung

cancer. Proteomics 3, 1678–1679.

47. Zhang, Y. F., Wu, D. L., Liu, W. W., et al. (2004)

Tree analysis of mass spectral urine profiles

discriminates transitional cell carcinoma of the

bladder from non cancer patient. Clin.

Biochem. 37, 772–779.

48. Kang, X., Xu, Y., Wu, X., et al. (2005) Proteomic

fingerprints for potential application to early

diagnosis of severe acute respiratory syn-

drome. Clin. Chem. 51, 56–64.

49. Bishop, C. M. (1995) Neural Networks for Pat-

tern Recognition. Oxford University Press,

Oxford, UK .

50. Rumelhart, D., Hinton, G., and Williams, R.

(1988) Learning internal representations by

error propagation. In: Neurocomputing, (Ander-

son, J. and Rosenfeld, E.), MIT Press, Cam-

bridge, MA, pp. 675–695.

51. Mian, S., Ball, G., Hornbuckle, J., et al. (2003)

Aprototype methodology combining surface

enhanced laser desorption/ionization protein

chip technology and artificial neural network

algorithms to predict the chemoresponsive-

ness of breast cancer cell lines exposed to

Paclitaxel and Doxorubicin under in vitro

condition. Proteomics 3, 1725–1737.

52. Ball, G., Mian, S., Allibone, R. O., et al. (2002)

An integrated approach using artificial neural

networks and SELDI mass spectrometry for

the classification of human tumors and rapid

identification of potential biomarkers. Bioin-

formatics 18, 395–404.

53. Poon, T. C. W., Yip, T., Chan, A. T. C., Yip, C.,

Yip, V., and Mok, T. S. K. (2003) Comprehen-

sive proteomic profiling identifies serum

proteomic signatures for detection of hepato-

cellular carcinoma and its subtypes. Clin.

Chem. 49, 752–760.

54. Kohonen, T. (1995) Self Organizing Maps.

Springer Publishers, Berlin, Germany.

55. Breiman, L. (1996) Bagging predictors. Machine

Learning 24, 123–140.

03_Jortani.qxd 11/21/06 3:07 PM Page 31

Clinical Proteomics________________________________________________________________ Volume 2, 2006

32 _______________________________________________________________________________ Thomas et al.

56. Izmirilan, G. (2004) Application of random

forest classification algorithm to a SELDI-TOF

Proteomics study in the setting of a cancer

prevention trial. Ann. NY Acad. Sci. 1020,

154–174.

57. Tourassi, G. D. and Floyd, C. E. (1997) The

effect of data sampling on the performance

evaluation of artificial neural networks in

medical diagnosis. Med. Dec. Mak. 17, 186–192.

58. Wilson, L. L., Tran, L., Morton, D. L., and

Hoon, D. S. B. (2004) Detection of Differentially

expressed proteins in early-stage melanoma

patients using SELDI-TOF mass spectrometry.

Ann. NY Acad. Sci. 1022, 317–322.

59. Tatay, J. W., Feng, X., Sobczak, N., et al. (2003)

Multiple approaches to data mining of pro-

teomic data based on statistical and pattern

classification methods. Proteomics 3, 1704–1709.

03_Jortani.qxd 11/21/06 3:07 PM Page 32

Download
Save PDF to desktop

Email
Email completed PDF

Send for Signing
Email for others to sign

Print
Print document

NAME

View Audit Log
Print With Audit Log
Duplicate Document

Fill Online, Printable, Fillable, Blank Original Article Clinical Proteomics Copyright 2006 Humana Press Form

Related forms

Fillable Original Article Clinical Proteomics Copyright 2006 Humana Press

Fill Online, Printable, Fillable, Blank Original Article Clinical Proteomics Copyright 2006 Humana Press Form

Related forms

First 20 documents completely FREE!