Difference between revisions of "Kevin Wyllie Week 14"

Revision as of 05:37, 7 December 2015

Electronic Lab Notebook

Statistical Analysis and Formatting

We looked at Dr. Dahlquist's comments on our prior worksheet to make a second attempt at processing and formatting for GenMAPP. Primarily, our biggest error in the the previous attempt was forgetting to account for the fact that each gene was spotted in (technical) quadruplicates.

1. Dr. Dahlquist prepared a sheet for us to start with, which separated each quadruplicate of the genome across columns. So the ID column for the first technical replicate was in column A, followed by fold change data for one of the four technical replicates within each of the biological replicates, for the (untreated) biofilm and tobramycin-treated biofilm samples. Next was the second ID column (identical to the first) in column J, followed by the second set of technical replicates within each biological replicate. These sheet was named "quadruplicated_spots_separated".
2. After all four sets of technical replicates, yet another ID column (column AK) was added, following by the averaged fold change values across each technical replicate (for each biological replicate - columns AL through AS).
   • For these averages the =AVERAGE(*four technical replicates*) command was used and pasted through the entirety of each column. The *four technical replicates* is a placeholder for the cell coordinates corresponding to each technical replicate within a biological replicate.
   • Because all five untreated and all three treated samples will be averaged later in the protocol, the word "consolidated" (instead of "average") was used in the headings for these columns. For example, the column heading for the first biological untreated sample was "consolidated_Biofilm_1_scaled_centered_4".
3. Next, a new sheet was created, named "statistics". Pasted into this sheet (using the "Paste Values" function) were the "consolidated" columns (preceded by the ID column). Thus, the ID's were in column A while the consolidated fold changes for each biological replicate were in columns B through J. The average across the five untreated samples were calculated in column K (using the previously mentioned command), with the header "AVG_Biofilm_scaled_centered". In column L, the same was done using the three treated samples.
4. In column M, the ratios between the averages for treated and untreated samples were calculated. However, because these fold changes are in log space the fold changes for the treated samples were subtracted from that of the untreated samples.
   • An example command for the gene in 2, is =L2-K2.
5. Next, the p-value's for each fold change ratio was calculated. A type-3, 2-tailed T-test was used.
   • The command for the first gene (row 2) is =TTEST(C2:G2,H2:J2,2,3). As before, this was pasted through all genes.
6. The next sheet was named "bonferroni_pval".

Sanity Check

There are 7251 genes in the sheet.

• How many genes have p value < 0.05? And what is the percentage?
  • 4318, ~60%
• What about p < 0.01?
  • 2971, ~41%
• What about p < 0.001?
  • 1460, ~20%
• What about p < 0.0001?
  • 645, ~9%

• How many genes are p < 0.05 for the Bonferroni-corrected p value?
  • 179, ~2.4%
• How many genes are p < 0.05 for the Benjamini and Hochberg-corrected p value?
  • 609, ~8.4%

• "Pvalue" < 0.05, and "Biofilm_Tobramycin_ratio" > 0.
• "Pvalue" < 0.05, and "Biofilm_Tobramycin_ratio" > 0.

This is a more realistic values for the fold change cut-offs because it represents about a 20% fold change which is about the level of detection of this technology:

• "Biofilm_Tobramycin_ratio" > 0.25 and "Pvalue" < 0.05.
• "Biofilm_tobramycin_ratio" < -0.25 and "Pvalue" < 0.05.

Links

User:kwyllie

• Kevin Wyllie Week 2 (See the original assignment and class journal.)
• Kevin Wyllie Week 3 (See the original assignment and class journal.)
• Kevin Wyllie Week 4 (See the original assignment and class journal.)
• Kevin Wyllie Week 5 (See the original assignment and class journal.)
• Kevin Wyllie Week 6 (See the original assignment and class journal.)
• Kevin Wyllie Week 7 (See the original assignment and class journal.)
• Kevin Wyllie Week 8 (See the original assignment and class journal.)
• Kevin Wyllie Week 9 (See the original assignment and class journal.)
• Kevin Wyllie Week 10 (See the original assignment.)
• Kevin Wyllie Week 11 (See the original assignment.)
• Kevin Wyllie Week 12 (See the original assignment.)
• Kevin Wyllie Week 14 (See the original assignment.)
• Kevin Wyllie Week 15 (See the original assignment.)