Open Access Research article

Relation between smoking history and gene expression profiles in lung adenocarcinomas

Johan Staaf12, Göran Jönsson12, Mats Jönsson1, Anna Karlsson1, Sofi Isaksson1, Annette Salomonsson1, Helen M Pettersson3, Maria Soller4, Sven-Börje Ewers1, Leif Johansson5, Per Jönsson6 and Maria Planck1*

Author Affiliations

1 Department of Oncology, Clinical Sciences, Lund University and Skåne University Hospital, Barngatan 2:1, SE-22185, Lund, Sweden

2 CREATE Health Strategic Center for Translational Cancer Research, Lund University, BMC C13, SE 221 84, Lund, Sweden

3 Center for Molecular Pathology, Department of Laboratory Medicine, Lund University, SE 20502, Malmö, Sweden

4 Department of Clinical Genetics, Lund University and Regional Laboratories Region Skåne, SE 22185, Lund, Sweden

5 Department of Pathology, Lund University and Regional Laboratories Region Skåne, SE 22185, Lund, Sweden

6 Department of Thoracic Surgery, Clinical Sciences, Lund University and Skåne University Hospital, SE 22185, Lund, Sweden

For all author emails, please log on.

BMC Medical Genomics 2012, 5:22  doi:10.1186/1755-8794-5-22

Published: 7 June 2012

Additional files

Additional file 1:

Table S1. AC1/AC2 and supervised gene expression centroids. An Excel file, Table S1, containing gene expression centroids for the AC1/AC2 and seven gene signatures derived from supervised analysis.

Format: XLS Size: 387KB Download file

This file can be viewed with: Microsoft Excel Viewer

Open Data

Additional file 2:

Table S2. Fraction of smokers, subdivided also into current and former status, classified as never-smokers by classifiers derived from supervised analysis of seven AC data sets. An Excel file, Table S2, describing the fraction of true smokers overall, current smokers, and former smokers classified as never-smokers by classifiers derived from supervised analysis of seven AC data sets.

Format: XLS Size: 27KB Download file

This file can be viewed with: Microsoft Excel Viewer

Open Data

Additional file 3:

Figure S1. Sensitivity and specificity of the DCC derived classifier for identification of never-smokers across different correlation classification cut-offs. A pdf file, Figure S1, showing the sensitivity and specificity of the DCC derived classifier for identification of never-smokers across different correlation classification cut-offs in seven data sets. Sensitivity and specificity for different Pearson correlation classification cut-offs are shown in the left subpanels, while the corresponding number of DCC-classified never-smokers and smokers are shown in the right panel for respective data set. A) DCC-classifier applied to the DCC data set. B) DCC-classifier applied to GSE10072. C) DCC-classifier applied to GSE12667. D) DCC-classifier applied to GSE11969. E) DCC-classifier applied to Beer et al. F) DCC-classifier applied to GSE32863. G) DCC-classifier applied to the original 39 adenocarcinomas.

Format: PDF Size: 121KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data

Additional file 4:

Table S3. Functional analysis of AC1/AC2 gene signature derived from unsupervised analysis, and the classifier derived from supervised analysis of the DCC data set using LitVAn and IPA. An Excel file, Table S3, showing results from Functional analysis of AC1/AC2 gene signature derived from unsupervised analysis, and the classifier derived from supervised analysis of the DCC data set using LitVAn and IPA.

Format: XLS Size: 22KB Download file

This file can be viewed with: Microsoft Excel Viewer

Open Data

Additional file 5:

Figure S2 Expression of the CIN70 metagene across seven AC data sets classified by both unsupervised and supervised analysis. A pdf file, Figure S2, showing the log2ratio expression of the CIN70 metagene across seven AC data sets classified by both unsupervised and supervised analysis. CIN70 metagene expression displayed as box plots for true never-smokers and smokers (white), true current, former and never-smokers (gray), AC1 and AC2 classified samples (blue), and DCC centroid classified samples (red) in A) the DCC data set, B) GSE10072, C) GSE11969, D) GSE12667, E) Beer et al., F) GSE32863, and G) the original Illumina cohort of 39 AC. P-values were calculated using Wilcoxon’s test (two groups) or Kruskal-Wallis test (three groups).

Format: PDF Size: 126KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data

Additional file 6:

Figure S3. Pack-year analysis of five data sets comprising normal airway epithelial cells or normal adjacent lung tissue classified by both unsupervised and supervised analysis. A pdf file, Figure S3, showing pack-year distribution for classification of five data sets using classifiers from unsupervised and supervised analyses. Pack-years for AC1/AC2 classification (A) or DCC-classification (B) for GSE7895, GSE11952, GSE19027, GSE19667 and GSE32863 respectively. P-values calculated using either Student’s t-test or Wilcoxon’s test.

Format: PDF Size: 118KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data

Additional file 7:

Figure S4. Expression of the CIN70 metagene across six data sets comprising normal airway epithelial cells or normal adjacent lung tissue classified by both unsupervised and supervised analysis. A pdf file, Figure S4, showing the expression of the CIN70 metagene across six data sets comprising normal airway epithelial cells or normal adjacent lung tissue classified by both unsupervised and supervised analysis. CIN70 metagene log2ratio expression are displayed as box plots for true never-smokers and smokers (white), true current, former and never-smokers (gray), AC1 and AC2 classified samples (blue), and DCC centroid classified samples (red) in A) GSE7895, B) GSE19027, C) GSE19667, D) GSE11952, E) normal samples in GSE10072, and F) normal samples in GSE32863. P-values were calculated using Wilcoxon’s test (two groups) or Kruskal-Wallis test (three groups).

Format: PDF Size: 129KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data