Kmill104 Week 9 - Revision history

Kmill104: /* Calculate the Benjamini & Hochberg p value Correction */ adding another step

2024-04-02T06:00:24Z

‎Calculate the Benjamini & Hochberg p value Correction: adding another step

Kmill104: /* Sanity Check: Number of genes significantly changed */ fixing bullet points

2024-04-02T05:58:55Z

‎Sanity Check: Number of genes significantly changed: fixing bullet points

Kmill104: /* Sanity Check: Number of genes significantly changed */ fixing tenses for this section

2024-04-02T05:54:02Z

‎Sanity Check: Number of genes significantly changed: fixing tenses for this section

Kmill104: /* Calculate the Benjamini & Hochberg p value Correction */ fixing tenses for this section

2024-04-02T05:25:54Z

‎Calculate the Benjamini & Hochberg p value Correction: fixing tenses for this section

Kmill104: /* Calculate the Bonferroni and p value Correction */ fixing tenses for this section

2024-04-02T05:17:11Z

‎Calculate the Bonferroni and p value Correction: fixing tenses for this section

Kmill104: /* Statistical Analysis Part 1: ANOVA */ updating tenses again

2024-04-02T05:12:58Z

‎Statistical Analysis Part 1: ANOVA: updating tenses again

Kmill104: /* Acknowledgements */ adding more detail

2024-03-26T06:34:50Z

‎Acknowledgements: adding more detail

Kmill104: /* Acknowledgements */ updating acknowledgements

2024-03-26T06:32:26Z

‎Acknowledgements: updating acknowledgements

Kmill104: /* Methods/Results */ updating steps

2024-03-26T06:30:19Z

‎Methods/Results: updating steps

Kmill104: /* Experimental Design and Getting Ready */ adding more detail

2024-03-26T05:54:21Z

‎Experimental Design and Getting Ready: adding more detail

@@ Line 63: / Line 63: @@
 # I then selected columns A through G, and sorted them by my MasterIndex in Column A in ascending order.
 # Finally, I copied column G and use Paste special > Paste values to paste it into the next column on the right of my ANOVA sheet.
 ==== Sanity Check: Number of genes significantly changed ====

@@ Line 77: / Line 77: @@
 We know that the Bonferroni correction is very stringent, and the Benjamini-Hochberg correction is less stringent.  To see this relationship, I filtered my data in the first Bonferroni column by clicking on the drop-down arrow and setting a criterion so that the p-value is less than 0.5. I then repeated this for the Benjamini-Hochberg column. I used this to answer the following questions.
-** '''''How many genes are p < 0.05 for the Bonferroni-corrected p value? and what is the percentage (out of 6189)?''''' 248 genes are p < 0.05. This is 4.01% of the total genes in the dataset.
+*'''''How many genes are p < 0.05 for the Bonferroni-corrected p value? and what is the percentage (out of 6189)?''''' 248 genes are p < 0.05. This is 4.01% of the total genes in the dataset.
-** '''''How many genes are p < 0.05 for the Benjamini and Hochberg-corrected p value? and what is the percentage (out of 6189)?''''' 1822 genes are p < 0.05. This is 29.44% of the total genes in the dataset.
+*'''''How many genes are p < 0.05 for the Benjamini and Hochberg-corrected p value? and what is the percentage (out of 6189)?''''' 1822 genes are p < 0.05. This is 29.44% of the total genes in the dataset.
-* I then organized these numbers into a table in a PowerPoint slide modeled off the sample slide posted in the Week 9 Assignment, and uploaded this slide to the wiki.
+I then organized these numbers into a table in a PowerPoint slide modeled off the sample slide posted in the Week 9 Assignment, and uploaded this slide to the wiki.
-* We know that the expression of the gene ''NSR1'' (ID: YGR159C)is known to be induced by cold shock. I found NSR1 by using the drop-down arrow on column C to search its name. and I then used this to answer the following questions. What is its unadjusted, Bonferroni-corrected, and B-H-corrected p values?
 **Unadjusted: 2.86939E-10
 **Bonferroni-corrected: 1.77586E-06
@@ Line 94: / Line 96: @@
 **Yes, the unadjusted, Bonferroni-corrected, and B-H-corrected p-values are all less than 0.05, indicating that NSR1 changes expression due to cold shock in this experiment.
-* I then found MSN1 by using the drop-down arrow for column C and searching its name, and then used this to answer the following questions.  '''''What is its unadjusted, Bonferroni-corrected, and B-H-corrected p values?
+I then found MSN1 by using the drop-down arrow for column C and searching its name, and then used this to answer the following questions.
 **Unadjusted: 0.563798852
 **Bonferroni-corrected: 3489.351093

@@ Line 66: / Line 66: @@
 ==== Sanity Check: Number of genes significantly changed ====
-Before we move on to further analysis of the data, we want to perform a more extensive sanity check to make sure that we performed our data analysis correctly.  We are going to find out the number of genes that are significantly changed at various p value cut-offs.
+I then performed another sanity check to make sure the data analysis was performed correctly by finding the number of genes that are significantly changed at various p-value cut-offs.
-* Go to your wt_ANOVA worksheet.
+* First, I went to my wt_ANOVA worksheet.
-* Select row 1 (the row with your column headers) and select the menu item Data > Filter > Autofilter (The funnel icon on the Data tab).  Little drop-down arrows should appear at the top of each column.  This will enable us to filter the data according to criteria we set.
+* I then selected row 1 and selected the menu item Data > Filter > Autofilter, causing drop-down arrows to appear.
-* Click on the drop-down arrow for the unadjusted p value.  Set a criterion that will filter your data so that the p value has to be less than 0.05.
+* I clicked on the drop-down arrow for the unadjusted p value, and set a criterion that filtered my data so that the p value has to be less than 0.05, and used this to follow the answering questions.
 ** '''''How many genes have p < 0.05?  and what is the percentage (out of 6189)?''''' 2528 genes have p < 0.05. This is 40.85% of the total genes in the dataset.
 ** '''''How many genes have p < 0.01? and what is the percentage (out of 6189)?''''' 1652 genes have p < 0.01. This is 26.69% of the total genes in the dataset.
@@ Line 76: / Line 76: @@
 ** '''''How many genes have p < 0.0001? and what is the percentage (out of 6189)?''''' 496 genes have p < 0.0001. This is 8.01% of the total genes in the dataset.
-* Note that it is a good idea to create a new worksheet in your workbook to record the answers to these questions.  Then you can write a formula in Excel to automatically calculate the percentage for you.
+We know that the Bonferroni correction is very stringent, and the Benjamini-Hochberg correction is less stringent.  To see this relationship, I filtered my data in the first Bonferroni column by clicking on the drop-down arrow and setting a criterion so that the p-value is less than 0.5. I then repeated this for the Benjamini-Hochberg column. I used this to answer the following questions.
 ** '''''How many genes are p < 0.05 for the Bonferroni-corrected p value? and what is the percentage (out of 6189)?''''' 248 genes are p < 0.05. This is 4.01% of the total genes in the dataset.
 ** '''''How many genes are p < 0.05 for the Benjamini and Hochberg-corrected p value? and what is the percentage (out of 6189)?''''' 1822 genes are p < 0.05. This is 29.44% of the total genes in the dataset.
-* In summary, the p value cut-off should not be thought of as some magical number at which data becomes "significant".  Instead, it is a moveable confidence level.  If we want to be very confident of our data, use a small p value cut-off.  If we are OK with being less confident about a gene expression change and want to include more genes in our analysis, we can use a larger p value cut-off.
-* We will compare the numbers we get between the wild type strain and the other strains studied, organized as a table.  Use this [[Media:BIOL367_S24_sample_p-value_slide.pptx | sample PowerPoint slide]] to see how your table should be formatted. '''''Upload your slide to the wiki.'''''
+* I then organized these numbers into a table in a PowerPoint slide modeled off the sample slide posted in the Week 9 Assignment, and uploaded this slide to the wiki.
-** Note that since the wild type data is being analyzed by one of the groups in the class, it will be sufficient for this week to supply just the data for your strain.  We will do the comparison with wild type at a later date.
+* We know that the expression of the gene ''NSR1'' (ID: YGR159C)is known to be induced by cold shock. I found NSR1 by using the drop-down arrow on column C to search its name. and I then used this to answer the following questions. What is its unadjusted, Bonferroni-corrected, and B-H-corrected p values?
 **Unadjusted: 2.86939E-10
 **Bonferroni-corrected: 1.77586E-06
@@ Line 94: / Line 91: @@
 **Time point 90: -2.04985
 **Time point 120: -0.60622
-*Note that the average Log fold change is what we called "wt_AvgLogFC_(TIME)" in step 3 of the ANOVA analysis. '''''Does ''NSR1'' change expression due to cold shock in this experiment? '''''
+*'''''Does ''NSR1'' change expression due to cold shock in this experiment? '''''
 **Yes, the unadjusted, Bonferroni-corrected, and B-H-corrected p-values are all less than 0.05, indicating that NSR1 changes expression due to cold shock in this experiment.
-* For fun, find MSN1 (from your [[Week 3]] assignment) in the dataset.  '''''What is its unadjusted, Bonferroni-corrected, and B-H-corrected p values?
+* I then found MSN1 by using the drop-down arrow for column C and searching its name, and then used this to answer the following questions.  '''''What is its unadjusted, Bonferroni-corrected, and B-H-corrected p values?
 **Unadjusted: 0.563798852
 **Bonferroni-corrected: 3489.351093

@@ Line 53: / Line 53: @@
 ==== Calculate the Benjamini & Hochberg p value Correction ====
-# Insert a new worksheet named "wt_ANOVA_B-H".
+# I then inserted a new worksheet named "wt_ANOVA_B-H".
-# Copy and paste the "MasterIndex", "ID", and "Standard Name" columns from your previous worksheet into the first two columns of the new worksheet.
+# I copy and pasted the "MasterIndex", "ID", and "Standard Name" columns from my previous worksheet into the first two columns of the new worksheet.
-# For the following, use Paste special > Paste values.  Copy your unadjusted p values from your ANOVA worksheet and paste it into Column D.
+# I used Paste special > Paste values for the following: I copied my unadjusted p values from my ANOVA worksheet and pasted it into Column D.
-# Select all of columns A, B, C, and D. Sort by ascending values on Column D. Click the sort button from A to Z on the toolbar, in the window that appears, sort by column D, smallest to largest.
+# I then selected all of columns A, B, C, and D. I sorted by ascending values on column D by selecting the sort button from A to Z on the toolbar, and in the window that appeared, clicked sort by column D, smallest to largest.
-# Type the header "Rank" in cell E1.  We will create a series of numbers in ascending order from 1 to 6189 in this column.  This is the p value rank, smallest to largest.  Type "1" into cell E2 and "2" into cell E3. Select both cells E2 and E3. Double-click on the plus sign on the lower right-hand corner of your selection to fill the column with a series of numbers from 1 to 6189.
+# I then typed the header "Rank" in cell E1. This column was used to create a series of numbers in ascending order from 1 to 6189. This gave me the p value rank, smallest to largest.  To assign ranks, I typed "1" into cell E2 and "2" into cell E3. I selected both cells E2 and E3, and then double-clicked on the plus sign on the lower right-hand corner of my selection to fill the column with a series of numbers from 1 to 6189.
-# Now you can calculate the Benjamini and Hochberg p value correction. Type wt_B-H_p-value in cell F1. Type the following formula in cell F2: <code>=(D2*6189)/E2</code> and press enter. Copy that equation to the entire column.
+# I used the following to calculate the Benjamini and Hochberg p value corrections. I typed wt_B-H_p-value in cell F1, and typed the following formula in cell F2: <code>=(D2*6189)/E2</code> and pressed enter. I then copied that equation to the entire column.
-# Type "wt_B-H_p-value" into cell G1.
+# I then typed "wt_B-H_p-value" into cell G1.
-# Type the following formula into cell G2: <code>=IF(F2>1,1,F2)</code> and press enter. Copy that equation to the entire column.
+# In G2, I typed the following formula: <code>=IF(F2>1,1,F2)</code> and pressed enter, then copied that equation to the entire column.
-# Select columns A through G.  Now sort them by your MasterIndex in Column A in ascending order.
+# I then selected columns A through G, and sorted them by my MasterIndex in Column A in ascending order.
-# Copy column G and use Paste special > Paste values to paste it into the next column on the right of your ANOVA sheet.
+# Finally, I copied column G and use Paste special > Paste values to paste it into the next column on the right of my ANOVA sheet.
 ==== Sanity Check: Number of genes significantly changed ====

@@ Line 47: / Line 47: @@
 ==== Calculate the Bonferroni and p value Correction ====
-''Note: Be sure to undo any filters that you have applied before continuing with the next steps.''
+# Then, I performed adjustments to the p value to correct for the [https://xkcd.com/882/ multiple testing problem].  I labeled the next two columns to the right of wt_p-value with the same label, wt_Bonferroni_p-value.
-# Now we will perform adjustments to the p value to correct for the [https://xkcd.com/882/ multiple testing problem].  Label the next two columns to the right with the same label, wt_Bonferroni_p-value.
+# In the first cell of the first column, I typed the equation <code>=AN2*6189</code>, and upon completion of this single computation, copied the formula throughout the entire column.
-# Type the equation <code>=<wt_p-value>*6189</code>, Upon completion of this single computation, use the Step (10) trick to copy the formula throughout the column.
+# I then replaced any corrected p value that was greater than 1 with the number 1 by typing the following formula into the first cell below the second wt_Bonferroni_p-value column: <code>=IF(AO2>1,1,AO2)</code>, and then copied the formula throughout the column.
 ==== Calculate the Benjamini & Hochberg p value Correction ====

@@ Line 26: / Line 26: @@
 # In the next empty column to the right of wt_ss_HO, I created the column headers wt_ss_(TIME) as in (3).
 # I counted that there are 23 data points for each time point of the wt strain data.
-# In the first cell below the header wt_ss_t15, I then typed <code>=SUMSQ(<range of cells for logFC_t15>)-COUNTA(<range of cells for logFC_t15>)*<AvgLogFC_t15>^2</code> and hit enter.
+# In the first cell below the header wt_ss_t15, I then typed <code>=SUMSQ(D2:G2)-COUNTA(D2:G2)*<AA2>^2</code> and hit enter.
-#* The <code>COUNTA</code> function counts the number of cells in the specified range that have data in them (i.e., does not count cells with missing values).
+#* Upon completion of this single computation, I then used the Step (7) trick to copy the formula throughout the column.
-#* The phrase <range of cells for logFC_t15> should be replaced by the data range associated with t15.
+# I then repeated this computation for the t30 through t120 data points.
-#* The phrase <AvgLogFC_t15> should be replaced by the cell number in which you computed the AvgLogFC for t15, and the "^2" squares that value.
+#*Below wt_ss_t30, I typed <code>=SUMSQ(H2:L2)-COUNTA(H2:L2)*<AB2>^2</code> and hit enter, then copied the formula throughout the column.
-#* Upon completion of this single computation, use the Step (7) trick to copy the formula throughout the column.
+#*Below wt_ss_t60, I typed <code>=SUMSQ(M2:P2)-COUNTA(M2:P2)*<AC2>^2</code> and hit enter, then copied the formula throughout the column.
-# Repeat this computation for the t30 through t120 data points.  Again, be sure to get the data for each time point, type the right number of data points, and get the average from the appropriate cell for each time point, and copy the formula to the whole column for each computation.
+#*Below wt_ss_t90, I typed <code>=SUMSQ(Q2:U2)-COUNTA(Q2:U2)*<AD>^2</code> and hit enter, then copied the formula throughout the column.
-# In the first column to the right of wt_ss_t120, create the column header wt_SS_full.
+#*Below wt_ss_t120, I typed <code>=SUMSQ(V2:Z2)-COUNTA(V2:Z2)*<AE2>^2</code> and hit enter, then copied the formula throughout the column.
-# In the first row below this header, type <code>=sum(<range of cells containing "ss" for each timepoint>)</code> and hit enter.
+# In the first column to the right of wt_ss_t120, I then created the column header wt_ss_full.
-# In the next two columns to the right, create the headers wt_Fstat and wt_p-value.
+# In the first row below this header, I typed <code>=sum(AG2:AK2)</code> and hit enter.
-# Recall the number of data points from (13) -> 23.
+# In the next two columns to the right, I then created create the headers wt_Fstat and wt_p-value.
-# In the first cell of the wt_Fstat column, type <code>=(23-5)/5)*(<(wt_ss_HO>-<wt_SS_full>)/<wt_SS_full></code> and hit enter.
+# There are 23 data points for each time point, which I used in the following formula.
-#Use the number from (13), which is 23. Also note that "5" is the number of timepoints.
+# In the first cell of the wt_Fstat column, I typed <code>=(23-5)/5)*(AF2-AL2)/AL2</code> and hit enter.
-#* Replace the phrase wt_ss_HO with the cell designation.
+#* I then copied this to the whole column.
-#* Replace the phrase wt_ss_full with the cell designation.
+# In the first cell below the wt_p-value header, I typed <code>=FDIST(AM2,5,23-5)</code>, and then copied this to the whole column.
-#* Copy to the whole column.
+# Before moving onto the next step, I performed a quick sanity check to see if the computations were done correctly.
-# In the first cell below the wt_p-value header, type <code>=FDIST(<wt_Fstat>,5,23-5)</code> replacing the phrase wt_Fstat> with the cell designation. Copy to the whole column.
+#* First, I clicked on cell A1 and clicked on the Data tab.  I then selected the Filter icon, and used the drop-down arrows that appeared to filter the data.
-# Before we move on to the next step, we will perform a quick sanity check to see if we did all of these computations correctly.
+#* I clicked on the drop-down arrow on my wt_p-value column, then selected "Number Filters". In the window that appeared, I set a criterion that filtered my data so that the p value had to be less than 0.05.
-#*  Click on cell A1 and click on the Data tab.  Select the Filter icon (looks like a funnel). Little drop-down arrows should appear at the top of each column. This will enable us to filter the data according to criteria we set.
+#* Excel then only displayed the rows that corresponded to data meeting that filtering criterion.  A number appeared in the lower left hand corner of the window giving me the number of rows that meet that criterion.  These numbers were then used to check our results with each other to make sure that the computations were performed correctly.
 ==== Calculate the Bonferroni and p value Correction ====

@@ Line 127: / Line 127: @@
 ==Acknowledgements==
-I worked under the guidance of Dr. Dahlquist on 3-14-24 and 3-18-24. I consulted with my classmates during class time when I had a question about 21. in the ANOVA section of this journal assignment. I first copied the procedure from the Week 9 Assignment page, and then adjusted the steps to what exactly I did when following the steps.
+I worked under the guidance of Dr. Dahlquist on 3-14-24 and 3-18-24. I consulted with my classmates during class time when I had a question about 21. in the ANOVA section of this journal assignment. I first copied the  general procedure from the Week 9 Assignment page, and then adjusted my individual procedure to what exactly I did when following the steps.
 Except for what is noted above, this individual journal entry was completed by me and not copied from another source.

@@ Line 9: / Line 9: @@
 * I then changed the filename so that it contained my initials to distinguish it from other students' work.'''
 * I then looked at the sheet labeled "Master_Sheet_wt" to find that:
-#* The strain that me and Dean will analyze is the wild type strain. The name of the original file is [[Media:BIOL367_S24_microarray-data_wt.xlsx | BIOL367_S24_microarray-data_wt.xlsx]]. The name of the file with my saved initials and what I will be working on is [[Media:BIOL367_S24_microarray-data_wt_KM.xlsx | BIOL367_S24_microarray-data_wt_KM.xlsx]]. For the total wt strain data, there are 23 replicates. At time point 15, there are 4 replicates. At time point 30, there are 5 replicates. At time point 60, there are 4 replicates. At time point 90, there are 5 replicates. And at time point 120, there are 5 replicates.
+** The strain that me and Dean will analyze is the wild type strain. The name of the original file is [[Media:BIOL367_S24_microarray-data_wt.xlsx | BIOL367_S24_microarray-data_wt.xlsx]]. The name of the file with my saved initials and what I will be working on is [[Media:BIOL367_S24_microarray-data_wt_KM.xlsx | BIOL367_S24_microarray-data_wt_KM.xlsx]]. For the total wt strain data, there are 23 replicates. At time point 15, there are 4 replicates. At time point 30, there are 5 replicates. At time point 60, there are 4 replicates. At time point 90, there are 5 replicates. And at time point 120, there are 5 replicates.
 ==== Statistical Analysis Part 1: ANOVA ====
-The purpose of the within-stain ANOVA test is to determine if any genes had a gene expression change that was significantly different than zero at '''''any''''' timepoint.
+# I then created a new worksheet, naming it "wt_ANOVA".
+# I then copied all data from the "Master_Sheet" worksheet and pasted it into my new worksheet.
-# Create a new worksheet, naming it "wt_ANOVA".
+# At the top of the first column to the right of your data, I then created five column headers of the form wt_AvgLogFC_(TIME) where (TIME) is 15, 30, 60, 90, and 120.
-# Copy all data from the "Master_Sheet" worksheet and paste it in your new worksheet.
+# In the cell below the wt_AvgLogFC_t15 header, I then typed <code>=AVERAGE(</code>
-# At the top of the first column to the right of your data, create five column headers of the form wt_AvgLogFC_(TIME) where wt is your strain designation and (TIME) is 15, 30, 60, 90, and 120.
+# I then highlighted all the data in row 2 associated with t15 (D1:G1), pressed the closing paren key (shift 0), and pressed the "enter" key.
-# In the cell below the wt_AvgLogFC_t15 header, type <code>=AVERAGE(</code>
+# The cell now contains the average of the log fold change data from the first gene at t=15 minutes.
-# Then highlight all the data in row 2 associated with t15, press the closing paren key (shift 0),and press the "enter" key.
+# I then clicked on this cell and positioned my cursor at the bottom right corner. After the cursor changed to a thin black plus sign, I double clicked and the formula was copied to the remaining genes.
-# This cell now contains the average of the log fold change data from the first gene at t=15 minutes.
+# I then repeated steps (4) through (8) with the t30 (H1:L1), t60 (M1:P1), t90 (Q1:U1), and the t120 data (V1:Z1).
-# Click on this cell and position your cursor at the bottom right corner. You should see your cursor change to a thin black plus sign (not a chubby white one). When it does, double click, and the formula will magically be copied to the entire column of 6188 other genes.
+# To the right of the wt_AvgLogFC_t120 calculation, I created the column header wt_ss_HO.
-# Repeat steps (4) through (8) with the t30, t60, t90, and the t120 data.
+# In the first cell below this header, I typed <code>=SUMSQ(</code>
-# Now in the first empty column to the right of the wt_AvgLogFC_t120 calculation, create the column header wt_ss_HO.
+# Then, I highlighted all the LogFC data in row 2 (D1:Z1), pressed the closing paren key (shift 0),and pressed the "enter" key.
-# In the first cell below this header, type <code>=SUMSQ(</code>
+# In the next empty column to the right of wt_ss_HO, I created the column headers wt_ss_(TIME) as in (3).
-# Highlight all the LogFC data in row 2 (but not the AvgLogFC), press the closing paren key (shift 0),and press the "enter" key.
+# I counted that there are 23 data points for each time point of the wt strain data.
-# In the next empty column to the right of wt_ss_HO, create the column headers wt_ss_(TIME) as in (3).
+# In the first cell below the header wt_ss_t15, I then typed <code>=SUMSQ(<range of cells for logFC_t15>)-COUNTA(<range of cells for logFC_t15>)*<AvgLogFC_t15>^2</code> and hit enter.
 #* The <code>COUNTA</code> function counts the number of cells in the specified range that have data in them (i.e., does not count cells with missing values).
 #* The phrase <range of cells for logFC_t15> should be replaced by the data range associated with t15.