Classification and Computational Methods in Gene Expression Data Analysis

Cecilia Ritz

Classification and Computational Methods in Gene Expression Data Analysis

Cecilia Ritz

Research output: Thesis › Doctoral Thesis (compilation)

Abstract

The technology of cDNA microarrays has given us the possibility to monitor the state of cells by measuring the activity of thousands of genes simultaneously. This high-throughput techniqe has in cancer research allowed exploratory studies of molecular mechanisms behind for example metastasis and response to therapy. This increased knowledge can hopefully result in new therapies and improved prognostic and predictive tools. These tools however have to be properly validated in large cohorts and must be subjected to large-scale trials before use in the clinic.

One aim of this thesis is to evaluate the performance of classifiers of clinical outcome for breast cancer based on gene expression data as compared to conventional clinical markers. Additionally, we develop computational methods for analysis and classification using gene expression data. Our results suggests that clinical markers and molecular profiling have similar power in breast cancer prognosis. Further studies using larger cohorts are thus needed to validate and refine molecular prognostic profiles. We have also performed multicategory classification of leukemia into genetic subtypes and have predicted response to therapy in a subgroup. The main contribution to the computational analysis is our development of a method for improvement of missing value imputation of 2-dye cDNA microarray data. Recognizing that some categories of missing values are over- or underestimated in a kNN-based imputation method, we suggest a linear model that corrects for this bias and improves imputation of these spots.

Original language	English
Qualification	Doctor
Awarding Institution
Supervisors/Advisors	Edén, Patrik, Supervisor
Award date	2007 May 11
Publisher	Department of Theoretical Physics, Lund University
ISBN (Print)	978-91-628-7159-8
Publication status	Published - 2007

Bibliographical note

Defence details

Date: 2007-05-11
Time: 10:15
Place: Lecture Hall F, Dept. of Physics

External reviewer(s)

Name: Caldas, Carlos
Title: Professor
Affiliation: Dept. of Oncology, Cambridge University, UK

---

<div class="article_info">Patrik Edén, Cecilia Ritz, Carsten Rose, Mårten Fernö and Carsten Peterson. <span class="article_issue_date">2004</span>. <span class="article_title">"Good old" clinical markers have similar power in breast cancer prognosis as microarray gene expression profilers</span> <span class="journal_series_title">European Journal of Cancer</span>, <span class="journal_volume">vol 40</span> <span class="journal_pages">pp 1837-1841</span>. <span class="journal_distributor">Elsevier</span></div>
<div class="article_info">Emma Niméus-Malmström, Cecilia Ritz, Patrik Edén, Anders Johnsson, Mattias Ohlsson, Carina Strand, Görel Östberg, Mårten Fernö and Carsten Peterson. <span class="article_issue_date">2006</span>. <span class="article_title">Gene expression profilers and conventional clinical markers to predict recurrences for premenopausal breast cancer patients after adjuvant chemotherapy</span> <span class="journal_series_title">European Journal of Cancer</span>, <span class="journal_volume">vol 42</span> <span class="journal_pages">pp 2729-2737</span>. <span class="journal_distributor">Elsevier</span></div>
<div class="article_info">Anna Andersson, Cecilia Ritz, David Lindgren, Patrik Edén, Carin Lassen, Jesper Heldrup, Tor Olofsson, Johan Råde, Magnus Fontes, Anna Porwit-MacDonald, Mikael Behrendtz, Mattias Höglund, Bertil Johansson and Thoas Fioretos. <span class="article_issue_date">2007</span>. <span class="article_title">Microarry-based classification of a consecutive series of 121 childhood acute leukemias: prediction of leukemic and genetic subtype as well as of minimal residual disease</span> <span class="journal_series_title">Leukemia</span>, <span class="journal_distributor">Nature publishing group</span> (inpress)</div>
<div class="article_info">Cecilia Ritz and Patrik Edén. <span class="article_issue_date"></span>. <span class="article_title">Missing value categorization improves imputation in 2-dye cDNA microarray data</span> (submitted)</div>

Subject classification (UKÄ)

Biophysics

Free keywords

Bioinformatik
medicinsk informatik
Bioinformatics
medical informatics
biomathematics biometrics
missing values
leukemia
cDNA microarray data
supervised classification
breast cancer
prognostic markers
biomatematik

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Cite this

@phdthesis{93d1cacaf48a4ca2a88d2827aabcf941,

title = "Classification and Computational Methods in Gene Expression Data Analysis",

abstract = "The technology of cDNA microarrays has given us the possibility to monitor the state of cells by measuring the activity of thousands of genes simultaneously. This high-throughput techniqe has in cancer research allowed exploratory studies of molecular mechanisms behind for example metastasis and response to therapy. This increased knowledge can hopefully result in new therapies and improved prognostic and predictive tools. These tools however have to be properly validated in large cohorts and must be subjected to large-scale trials before use in the clinic. One aim of this thesis is to evaluate the performance of classifiers of clinical outcome for breast cancer based on gene expression data as compared to conventional clinical markers. Additionally, we develop computational methods for analysis and classification using gene expression data. Our results suggests that clinical markers and molecular profiling have similar power in breast cancer prognosis. Further studies using larger cohorts are thus needed to validate and refine molecular prognostic profiles. We have also performed multicategory classification of leukemia into genetic subtypes and have predicted response to therapy in a subgroup. The main contribution to the computational analysis is our development of a method for improvement of missing value imputation of 2-dye cDNA microarray data. Recognizing that some categories of missing values are over- or underestimated in a kNN-based imputation method, we suggest a linear model that corrects for this bias and improves imputation of these spots.",

keywords = "Bioinformatik, medicinsk informatik, Bioinformatics, medical informatics, biomathematics biometrics, missing values, leukemia, cDNA microarray data, supervised classification, breast cancer, prognostic markers, biomatematik",

author = "Cecilia Ritz",

note = "Defence details Date: 2007-05-11 Time: 10:15 Place: Lecture Hall F, Dept. of Physics External reviewer(s) Name: Caldas, Carlos Title: Professor Affiliation: Dept. of Oncology, Cambridge University, UK --- <div class={"}article_info{"}>Patrik Ed{\'e}n, Cecilia Ritz, Carsten Rose, M{\aa}rten Fern{\"o} and Carsten Peterson. <span class={"}article_issue_date{"}>2004</span>. <span class={"}article_title{"}>{"}Good old{"} clinical markers have similar power in breast cancer prognosis as microarray gene expression profilers</span> <span class={"}journal_series_title{"}>European Journal of Cancer</span>, <span class={"}journal_volume{"}>vol 40</span> <span class={"}journal_pages{"}>pp 1837-1841</span>. <span class={"}journal_distributor{"}>Elsevier</span></div> <div class={"}article_info{"}>Emma Nim{\'e}us-Malmstr{\"o}m, Cecilia Ritz, Patrik Ed{\'e}n, Anders Johnsson, Mattias Ohlsson, Carina Strand, G{\"o}rel {\"O}stberg, M{\aa}rten Fern{\"o} and Carsten Peterson. <span class={"}article_issue_date{"}>2006</span>. <span class={"}article_title{"}>Gene expression profilers and conventional clinical markers to predict recurrences for premenopausal breast cancer patients after adjuvant chemotherapy</span> <span class={"}journal_series_title{"}>European Journal of Cancer</span>, <span class={"}journal_volume{"}>vol 42</span> <span class={"}journal_pages{"}>pp 2729-2737</span>. <span class={"}journal_distributor{"}>Elsevier</span></div> <div class={"}article_info{"}>Anna Andersson, Cecilia Ritz, David Lindgren, Patrik Ed{\'e}n, Carin Lassen, Jesper Heldrup, Tor Olofsson, Johan R{\aa}de, Magnus Fontes, Anna Porwit-MacDonald, Mikael Behrendtz, Mattias H{\"o}glund, Bertil Johansson and Thoas Fioretos. <span class={"}article_issue_date{"}>2007</span>. <span class={"}article_title{"}>Microarry-based classification of a consecutive series of 121 childhood acute leukemias: prediction of leukemic and genetic subtype as well as of minimal residual disease</span> <span class={"}journal_series_title{"}>Leukemia</span>, <span class={"}journal_distributor{"}>Nature publishing group</span> (inpress)</div> <div class={"}article_info{"}>Cecilia Ritz and Patrik Ed{\'e}n. <span class={"}article_issue_date{"}></span>. <span class={"}article_title{"}>Missing value categorization improves imputation in 2-dye cDNA microarray data</span> (submitted)</div>",

year = "2007",

language = "English",

isbn = "978-91-628-7159-8",

publisher = "Department of Theoretical Physics, Lund University",

type = "Doctoral Thesis (compilation)",

}

TY - THES

T1 - Classification and Computational Methods in Gene Expression Data Analysis

AU - Ritz, Cecilia

N1 - Defence details Date: 2007-05-11 Time: 10:15 Place: Lecture Hall F, Dept. of Physics External reviewer(s) Name: Caldas, Carlos Title: Professor Affiliation: Dept. of Oncology, Cambridge University, UK --- <div class="article_info">Patrik Edén, Cecilia Ritz, Carsten Rose, Mårten Fernö and Carsten Peterson. <span class="article_issue_date">2004</span>. <span class="article_title">"Good old" clinical markers have similar power in breast cancer prognosis as microarray gene expression profilers</span> <span class="journal_series_title">European Journal of Cancer</span>, <span class="journal_volume">vol 40</span> <span class="journal_pages">pp 1837-1841</span>. <span class="journal_distributor">Elsevier</span></div> <div class="article_info">Emma Niméus-Malmström, Cecilia Ritz, Patrik Edén, Anders Johnsson, Mattias Ohlsson, Carina Strand, Görel Östberg, Mårten Fernö and Carsten Peterson. <span class="article_issue_date">2006</span>. <span class="article_title">Gene expression profilers and conventional clinical markers to predict recurrences for premenopausal breast cancer patients after adjuvant chemotherapy</span> <span class="journal_series_title">European Journal of Cancer</span>, <span class="journal_volume">vol 42</span> <span class="journal_pages">pp 2729-2737</span>. <span class="journal_distributor">Elsevier</span></div> <div class="article_info">Anna Andersson, Cecilia Ritz, David Lindgren, Patrik Edén, Carin Lassen, Jesper Heldrup, Tor Olofsson, Johan Råde, Magnus Fontes, Anna Porwit-MacDonald, Mikael Behrendtz, Mattias Höglund, Bertil Johansson and Thoas Fioretos. <span class="article_issue_date">2007</span>. <span class="article_title">Microarry-based classification of a consecutive series of 121 childhood acute leukemias: prediction of leukemic and genetic subtype as well as of minimal residual disease</span> <span class="journal_series_title">Leukemia</span>, <span class="journal_distributor">Nature publishing group</span> (inpress)</div> <div class="article_info">Cecilia Ritz and Patrik Edén. <span class="article_issue_date"></span>. <span class="article_title">Missing value categorization improves imputation in 2-dye cDNA microarray data</span> (submitted)</div>

PY - 2007

Y1 - 2007

N2 - The technology of cDNA microarrays has given us the possibility to monitor the state of cells by measuring the activity of thousands of genes simultaneously. This high-throughput techniqe has in cancer research allowed exploratory studies of molecular mechanisms behind for example metastasis and response to therapy. This increased knowledge can hopefully result in new therapies and improved prognostic and predictive tools. These tools however have to be properly validated in large cohorts and must be subjected to large-scale trials before use in the clinic. One aim of this thesis is to evaluate the performance of classifiers of clinical outcome for breast cancer based on gene expression data as compared to conventional clinical markers. Additionally, we develop computational methods for analysis and classification using gene expression data. Our results suggests that clinical markers and molecular profiling have similar power in breast cancer prognosis. Further studies using larger cohorts are thus needed to validate and refine molecular prognostic profiles. We have also performed multicategory classification of leukemia into genetic subtypes and have predicted response to therapy in a subgroup. The main contribution to the computational analysis is our development of a method for improvement of missing value imputation of 2-dye cDNA microarray data. Recognizing that some categories of missing values are over- or underestimated in a kNN-based imputation method, we suggest a linear model that corrects for this bias and improves imputation of these spots.

AB - The technology of cDNA microarrays has given us the possibility to monitor the state of cells by measuring the activity of thousands of genes simultaneously. This high-throughput techniqe has in cancer research allowed exploratory studies of molecular mechanisms behind for example metastasis and response to therapy. This increased knowledge can hopefully result in new therapies and improved prognostic and predictive tools. These tools however have to be properly validated in large cohorts and must be subjected to large-scale trials before use in the clinic. One aim of this thesis is to evaluate the performance of classifiers of clinical outcome for breast cancer based on gene expression data as compared to conventional clinical markers. Additionally, we develop computational methods for analysis and classification using gene expression data. Our results suggests that clinical markers and molecular profiling have similar power in breast cancer prognosis. Further studies using larger cohorts are thus needed to validate and refine molecular prognostic profiles. We have also performed multicategory classification of leukemia into genetic subtypes and have predicted response to therapy in a subgroup. The main contribution to the computational analysis is our development of a method for improvement of missing value imputation of 2-dye cDNA microarray data. Recognizing that some categories of missing values are over- or underestimated in a kNN-based imputation method, we suggest a linear model that corrects for this bias and improves imputation of these spots.

KW - Bioinformatik

KW - medicinsk informatik

KW - Bioinformatics

KW - medical informatics

KW - biomathematics biometrics

KW - missing values

KW - leukemia

KW - cDNA microarray data

KW - supervised classification

KW - breast cancer

KW - prognostic markers

KW - biomatematik

M3 - Doctoral Thesis (compilation)

SN - 978-91-628-7159-8

PB - Department of Theoretical Physics, Lund University

ER -

Classification and Computational Methods in Gene Expression Data Analysis

Abstract

Bibliographical note

Subject classification (UKÄ)

Free keywords

UN SDGs

Fingerprint

Cite this