Lucia I. Ramirez Week 11

For the modeling project, we will analyze a Dahlquist lab microarray dataset comparing the wild type strain to a different strain of yeast. For the statistical analysis, I will analyze the wild type data and Lauren will analyze the alternate strain:

• Wild type vs. Δhmo1: Lucia and Lauren

Excell Spreadsheet File name: Dahlquist_Lab_Microarray_Data_wt

Experimental Design

In the Excel spreadsheet, there is a worksheet labled "data". In this worksheet, each row contains the data for one gene (one spot on the microarray). The first column (labeled "ID") contains the gene identifier from the Saccharomyces Genome Database. The second column contains the Standard Name for each of the genes. Each subsequent column contains the log₂ ratio of the red/green fluorescence from each microarray hybridized in the experiment (steps 1-5 above having been performed for you already).

Each of the column headings from the data begin with the experiment name ("wt" for wild type S. cerevisiae data, "dCIN5" for the Δcin5 data, etc., and Spar for the S. paradoxus data). "LogFC" stands for "Log₂ Fold Change" which is the Log₂ red/green ratio. The timepoints are designated as "t" followed by a number in minutes. Replicates are numbered as "-0", "-1", "-2", etc. after the timepoint.

The timepoints are t15, t30, t60 (cold shock at 13°C) and t90 and t120 (cold shock at 13°C followed by 30 or 60 minutes of recovery at 30°C).

Number of replicates for each strain:

t15 = 4

t30 = 5

t60 = 4

t90 = 5

t120 = 5

total number of data points = 23

Statistical Analysis Part 1: ANOVA

1. Created a new worksheet, naming it stats
2. Copied the first two columns of the data worksheet (containing ID and Standard Name) into the stats sheet.
3. In the first row, columns C through G, created column labels of the form (STRAIN)_xbar_(TIME) where (STRAIN) is wt, dGLN3, etc., and (TIME) is 15, 30, etc.
4. In the first row, columns H and I, created the column labels (STRAIN)_xbar_grand and (STRAIN)_ss_HO.
5. In the first row, columns J through N, created the column labels (STRAIN)_ss_(TIME) as in (3).
6. In the first row, columns O, P, and Q, created the column labels (STRAIN)_SS_full, Fstat and p-value.
7. Computed the following:
   1. In cell C2, typed =AVERAGE(
   2. Clicked on the tab containing the data, and highlighted all the data in row 2 associated with (STRAIN) and t15, pressed the closing paren key (shift 0),and pressed the "enter" key.
   3. Clicked on the tab for the stats sheet. Cell C2 now contains the average of the log fold change data from the first gene at t=15 minutes.
   4. To copy the entire column of 6188 other genes, clicked on cell c2 and positioned cursor at the bottom right corner; cursor changes to a thin black plus sign (not a chubby white one); When it does, double click, and the formula.
   5. Moved to cell D2, and repeated (7) through (10) with the t30 data, to E2 with the t60 data, F2 with the t90, G2 with the 120.
   6. Moved to cell H2, and repeated (7) through (10) highlighting all the data for (STRAIN) in row 2 instead of the individual time points.
8. Computed the following:
   1. In cell I2, typed =SUMSQ(
   2. Clicked on the data sheet's tab again, and highlight all the data in row 2 for your (STRAIN), press the closing paren key (shift 0),and press the "enter" key.
      • The data highlighted here will be same as in (12).
9. Computed the following for the wt_ss columns:
   1. In cell J2, typed =SUMSQ(data!C2:F2) - 4*C2^2 and hit enter.
   2. In cell K2, typed =SUMSQ(data!G2:K2) - 5*D2^2 and hit enter.
   3. In cell L2, typed =SUMSQ(data!L2:O2) - 4*E2^2 and hit enter.
   4. In cell M2, typed =SUMSQ(data!P2:T2) - 5*F2^2 and hit enter.
   5. In cell N2, typed =SUMSQ(data!U2:Y2) - 5*G2^2 and hit enter.
      • The phrase "data!C2:F2" should be the data associated with t15. The number "4" is the number of data points (note that cells c2, d2, e2, f2 contain 4 data points). The phrase "stats!c2" gets the average you computed in Step (8) for t15, and the "^2" squares that value. Upon completion of this single computation, use the Step (10) trick to copy the formula throughout the column.
10. Once I populated cells J2 through N2, click on o2 and type =sum(j2:n2) and hit enter. Copy to the whole column.
11. Recalled the number of data points, which is 23 and called that total n and computed the following:
    1. In cell P2, type =((n-5)/5)*(i2-o2)/o2 and hit enter. Copied to the whole column.
    2. In cell Q2, type =FDIST(P2,5,n-5). Copied to the whole column.
12. Performed adjustments to the p value to correct for the multiple testing problem. Labeled column R "STRAIN_Bonferroni_p-value".
13. Typed the equation =q2*6189. Copied the whole column.
14. Replaced any corrected p value that is greater than 1 by the number 1 by typing the following formula into cell s2: =IF(r2>1,1,r2)

How many genes have p < 0.05? and what is the percentage (out of 6189)? 2378 (~38%)

How many genes have p < 0.01? and what is the percentage (out of 6189)? 1527 (~25%)

How many genes have p < 0.001? and what is the percentage (out of 6189)? 860 (~14%)

How many genes have p < 0.0001? and what is the percentage (out of 6189)? 460 (~7%)

How many genes are p < 0.05 for the Bonferroni-corrected p value? and what is the percentage (out of 6189)? 228 (~4%)

How many genes are p < 0.05 for the Benjamini and Hochberg-corrected p value? and what is the percentage (out of 6189)? 228 (~4%)