Difference between revisions of "Mpetredi Week 8"

Revision as of 21:13, 10 October 2013

Mitchell Petredis

[[Team Name]]

Relevant Documents

• Excel Spreadsheet of Merrel Compiled Raw Data Vibrio (from class)
• Text Document of Merrel Compiled raw Data Vibrio (from class)

References

• Week 8
• [OpenWetWare Instructions]
• [Merrel's publication]

Lab Journal

• Note: all of these events occurred after the introductory lab on October 10, 2013. Any instructions prior to the ones below can be found in the sections above "Sanity Check: Number of genes significantly changed" at the OpenWetWare link above, which was followed word for word without any issues, with the exception of the equation "=TDIST(ABS(R2),degrees of freedom,2" under the section "Perform statistical analysis on the ratios" which was properly changed to "=TDIST(ABS(Q2),degrees of freedom,2" in order to fix an error message in Excel.

1. Open Excel Document and in Microsoft Excel 2010 for PC, look at the top menu bar and navigate to Data/Filter. Small arrows should now appear to the right of every cell on the top row.
2. Find the "Pvalue" column, click on the first cell labeled "Pvalue", and click on the small downward facing arrow to the right of the cell text. A list of options will open in a small window next to the cell.
   • Clicking on the "Number Filters" option will open a window adjacent to the option, with another list of options to choose from. Select "Custom Filter..."
   • A new dialog box appears with two scroll-down boxes and two circle checks; by default, the first scroll-down box is labeled "equals" with nothing appearing in the second scroll-down box, and the circle "And" is marked instead of "Or". Click on the "equals" box and change it to "is less than", and to the right of that type in "0.05" in the text box. #*Clicking "OK" will then only display Pvalues less than 0.05 and also changes the downward facing arrow icon to the right of "Pvalues" to a smaller downward facing arrow and a filter icon, indicating that the filter is active. Notice the information at the bottom left of the Excel program that tells us how many values match that of less than 0.05 (in this example, 948 of 5221 records fit the criteria we set).
     • Repeat the steps in #2 for Pvalues < 0.01, 0.001, and 0.0001.
       • Results for other Pvalues
         • <0.01; 235/5221
         • <0.001; 24/5221
         • <0.0001; 2/5221
3. "When we use a p value cut-off of p < 0.05, what we are saying is that you would have seen a gene expression change that deviates this far from zero less than 5% of the time."
   • Replace 0.05 with 0.01, 0.001, 0.0001 to get better certainty of data results.
4. Change Pvalue filter back to 0.05, and add a filter to AvgLogFC_all to display results greater than 0.
   • Results display 352/5221 records
5. Keep Pvalue filter set to 0.05, but change AvgLogFC_all filter to display results less than 0.
   • Results display 596/5221 records
6. Keep Pvalue filter set to 0.05, but change AvgLogFC_all filter to display results greater than 0.25.
   • Results display 339/5221 records
7. Keep Pvalue filter set to 0.05, but change AvgLogFC_all filter to display results less than -0.25
   • Results display 579/5221 records
8. Merrel used the Statistical Analysis for Microarrays (SAM) program and organized her data by determining which genes were deferentially regulated (237 genes), induced (44 genes), and repressed (193). I could not find any mention of Pvalues or other similar aspects of data throughout the paper, and therefore could not quite compare Merrel's results to my results.