Supplementary Information Guide
1. Titles and Summaries of SI files:
Supplementary Table 1. Genes that predict pathway deregulation. List of genes selected in supervised analyses to discriminate control cells expressing GFP from cells expressing the indicated oncogenic activity.
Supplementary Table 2. Ras mutation status in NSCLC samples. Lung tumors were screened for mutation at K-Ras codon 12 by artificial restriction fragment length polymorphism and sequencing as described in Methods. Tumors mutation status is identified as positive (y) or negative (n).
Supplementary Table 3. Characteristics of breast cancer cell lines. Pathway activation status in the breast cancer cell lines was performed as described in Materials and Methods. To quantify Western blot analyses, the average intensity value of each fixed area is measured. These values are presented as % relative to highest value.
Supplementary Figure 1. Biochemical assays of pathway activation. HMEC were infected with either control GFP or a specific oncogene following 36 hours of serum starvation. After 18 hours, cells were collected, and Western Blotting analysis was performed as described in Materials and Methods to measure the expression of the encoded protein or downstream targets of the pathway.
Supplementary Figure 2. Gene expression patterns that predict oncogenic pathway deregulation. Leave-one-out cross-validation predicted classification probabilities for each individual sample. Pathway status for each experimental sample was predicted using a model generated independently of that sample. These predictions are based on the screened subset of discriminatory genes that comprise each signature model. The values on the horizontal axis are estimates of the overall signature scores in the regression analysis, and the corresponding values on the vertical axis are estimated classification probabilities. The GFP control samples are shown in blue and the oncogenic pathway samples in red.
Supplementary Figure 3. Validation of pathway predictions in tumors. Relationship of Ras pathway status in NSCLC samples to cell type of tumor origin. Prediction of Ras status in tumors is presented as a colorbar, where samples were sorted from low (blue) to high (red) activity. The corresponding tumor cell type is indicated as either squamous (S) or adenocarcinoma (A). Ras mutation status indicated by (*).
Supplementary Figure 4. Kaplan-Meier survival analysis for cancer patients based on individual pathway predictions for the tumor dataset.
- Lung cancer. Patients were classified as low or high probability of activation of the indicated pathway based on expression signatures (low probability <50%; high probability 50%). Kaplan-Meier survival curves were then generated for these two groups.
- Breast cancer. Same as in panel A.
- Ovarian cancer. Same as in panel A.
Supplementary Figure 5. Assays for pathway activities in breast cancer cell lines. Activity of E2F3, Myc, Src, b-catenin, and H-Ras pathways was determined as described in Supplementary Materials and Methods.
Supplementary Figure 6. Relationship of drug sensitivity to predictions of untargeted pathways. The degree of proliferation inhibition was plotted as a function of pathway prediction not specific to the drug treatment.
Supplementary Materials and Methods. This file contains a more detailed materials and methods description.
2. Format type and size of SI files:
Sup.Table1_2005-04-04260B.pdf: Supplementary Table 1. PDF file, 89.3 KB
Sup.Table2_2005-04-04260B.pdf: Supplementary Table 2. PDF file, 39.7 KB
Sup.Table3_2005-04-04260B.pdf: Supplementary Table 3. PDF file, 53.0 KB
Sup.Fig-1_2005-04-04260B.pdf: Supplementary Figure 1. PDF file, 190 KB
Sup.Fig-2_2005-04-04260B.pdf: Supplementary Figure 2. PDF file, 1.08 MB
Sup.Fig-3_2005-04-04260B.pdf: Supplementary Figure 3. PDF file, 1.12 MB
Sup.Fig-4_2005-04-04260B.pdf: Supplementary Figure 4. PDF file, 8.17 MB
Sup.Fig-5_2005-04-04260B.pdf: Supplementary Figure 5. PDF file, 206 KB
Sup.Fig-6_2005-04-04260B.pdf: Supplementary Figure 6. PDF file, 1.08 MB
Supplemental Materials and Methods. 38.5 KB