Sarah Carratt: Week 12
|Line 60:||Line 60:|
Revision as of 00:39, 12 April 2011
Preparing to Use STEM
- First, I downloaded the software and registered with the website. STEM
- After unziping the file (7-zip > Extract Here), I launched the program using the command window.
- To do this, I went to the start menu and clicked Programs > Accessories > Command Prompt.
- Then I entered the following commands in the window that appeared:
java -mx512M -jar stem.jar -d defaults.txt
Preparing the Spreadsheet
- I opened my master spreadsheet and inserted a new worksheet and named it "stem".
- I copied over data from the "final" worksheet to the "stem" worksheet.
- Renamed the columns: "MasterIndex"→ "SPOT" and "ID"→ "Gene Symbol"
- Deleted all of the data columns except AvgLogFC columns.
- Renamed the data columns with time and units for simplicity.
- 'Save as Text (Tab-delimited) (*.txt).
- Expression Data Info: selected my file (no normalization/add 0 and Spot IDs included in the data file)
- Gene Info: 'Saccharomyces cerevisiae (SGD) with no cross references'and no gene locations
- Options: STEM Clustering Method was selected, no changes
- Execute: Run the program
Viewing and Saving STEM Results
- Changed to "Based on real time" from Interface Options and took a screenshot of this window (saved to powerpoint)
- Opened detailed plots of each profile and took individual screenshots (saved to powerpoint)
- "Profile Gene Table" and "Profile GO Table": saved tables and uploaded to lionshare with correct names
Analyzing and Interpreting STEM Results
- Selected profile 9 for further interpretation, which was mostly down regulated and never up-regulated. I chose this gene because I find down-regulated genes easier to understand and explain. It also followed a simple pattern that I knew I would be able to put into context.
- 221 genes were assigned to this profile.
- 55.9 (56) genes were expected to be assigned to this profile.
- The p value is 1.5E-65 (significant).
- Opened the GO list and selected the third row. From the menu, I clicked Data > Filter > Autofilter.
- Looked at terms with p value of < 0.05: # terms
- Looked at corrected values with p value of < 0.05: # terms
10 Gene Ontology terms from your filtered list (either p < 0.05 or corrected p < 0.05). Look up the definitions for each of the terms at http://geneontology.org. Write a paragraph that describes the biological interpretation of these GO terms. In other words, why does the cell react to cold shock by changing the expression of genes associated with these GO terms?
- Opened web window .
- Opened gene list in Excel for profile 9.
- Copied the list of gene IDs into web box for ORFs/Genes.
- Checked the box for Check for all TFs.
- Unchecked the box for Indirect Evidence.
- Clicked the Search button.
- Top 10 transcription factors: Ste12p (36.2%), Rap1p (29.9%), Fhl1p (18.6%), Cin5p (14.0%), Phd1p (13.6%), Sok2p (13.1%), Yap6p (12.7%), Yap5p (11.3%), Skn7p (10.4%), Yap1p (9.5%).
- GLN3 is on the list, representing 3.2% and 7 genes: YDR210w, YEL007w, UGA1, CPS1, PUT1, ZEO1, WTM1.
- Transcription factors I used to general the matrix and diagram: CIN5, CUP9, FHL1, GTS1, HSF1, MSN1, MSN4, NRG1, RAP1, RCS1, REB1, ROX1, RPH1, YAP1, YAP6, GLN3, STE12, PHD1, SOK2, YAP5, SKN7
- I added the top five (non-overlapping) transcription factors to the list because I figured that if they represent such a large portion of regulated genes that they should be included in the map.
- Before I generated the figures, I unchecked the box for "Indirect Evidence" and selected "JPEG" from the drop-down menu for the "Output Image".
- Clicked "Generate"
- Saved RegulationMatrix to Lionshare.
- Clicked on the "Image" link to see the diagram of the network.
- Pasted image into my PowerPoint file and uploaded to Lionshare.