Lung disease diagnosis from gene expression profiles

Ma, Shuyi

Permalink

https://hdl.handle.net/2142/24449 Copy

Description

Title

Lung disease diagnosis from gene expression profiles

Author(s)

Ma, Shuyi

Issue Date

2011-05-25T14:25:51Z

Director of Research (if dissertation) or Advisor (if thesis)

Price, Nathan D.

Department of Study

Chemical & Biomolecular Engr

Discipline

Chemical Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• lung disease
• lung cancer
• asthma
• Chronic obstructive pulmonary disease (COPD)
• disease diagnosis
• systems biology
• Top Scoring Pair algorithm

Abstract

Lung diseases include some of the most widespread and deadly conditions known to affect people in the US today. One of the main challenges in treating lung disease is the difficulty of diagnosis. Clinical diagnosis remains largely dependent upon symptomatic-based diagnoses; many cases can be either misdiagnosed or undiagnosed until disease has progressed to a more severe stage. Most studies aimed at finding molecular-based diagnostics have focused on one or two diseases at a time, yielding limited success. Instead, we searched for biomarkers reflective of the global health state of the lung by studying data taken from a broad range of lung diseases. We used gene expression microarray data from five different lung diseases—lung adenocarcinoma, lung squamous cell carcinoma, large cell lung carcinoma, chronic obstructive pulmonary disease, and asthma—as well as a non-diseased phenotype, to train a classification tree scheme based on the Top Scoring Pair algorithm (Geman et al., Stat Appl Genet Mol Biol. 2004; 3: Article 19). The algorithm identified a 32 gene-pair panel that classified all of the phenotypes considered and another panel of 21 gene pair classifiers that classified the three cancers explicitly with sensitivity of 67±8% and 79±6% in ten-fold cross validation (p < 0.001), respectively. Several of the markers have been previously cited in literature as linked to these cancers. Thus, a TSP-based classification tree scheme accurately identifies lung diseases from the relative expression of a parsimonious set of diagnostic gene pairs.

Graduation Semester

2011-05

Permalink

http://hdl.handle.net/2142/24449

Copyright and License Information

Copyright 2011 Shuyi Ma

Owning Collections