Alondra Vega: Week 11

Electronic Laboratory Notebook

 * There are four replicates for each of the timepoints in the the ΔGLN3 Dahlquist dataset.
 * We took the average and standard deviations of each column (time point and replicate) to be able to normalize the data.
 * Once normalized, we subtracted the average from the cell and divided by the standard deviation to all cells in the designated columns.
 * It is important to keep the same standard deviation and average the same when doing the scaling and centering because we want to be able to compare the data. By using the column average and standard deviation, we are able to scale and center the data by the same factor all throughout the column.  This will help in the statistical analysis.
 * There were no replacements made when doing the statistical analysis.
 * To perform statistical analysis, we made a new page and took the average of the each time point.
 * Then we looked at the T distribution and the p-values.
 * The degrees of freedom in this case were 3, since we have four replicates for each time point.

Checking the p-value

 * How many genes have p-value< 0.05?
 * Time point 15 min: 763 genes
 * Time point 30 min: 1539 genes
 * Time point 60 min: 1559 genes
 * Time point 90 min: 538 genes
 * Time point 120min: 564 genes


 * How many genes have p-value< 0.01?
 * Time point 15 min: 210 genes
 * Time point 30 min: 456 genes
 * Time point 60 min: 384 genes
 * Time point 90 min: 129 genes
 * Time point 120min: 114 genes


 * How many genes have a p-value< 0.001?
 * Time point 15 min: 19 genes
 * Time point 30 min: 55 genes
 * Time point 60 min: 51 genes
 * Time point 90 min: 9 genes
 * Time point 120min: 16 genes


 * How many genes have a p-value<0.0001?
 * Time point 15 min: 1 gene
 * Time point 30 min: 2 genes
 * Time point 60 min: 6 genes
 * Time point 90 min: 3 genes
 * Time point 120min: 5 genes


 * Bonferroni Correction when p<0.05
 * Time point 15 min: 1 gene
 * Time point 30 min: 0 genes
 * Time point 60 min: 2 genes
 * Time point 90 min: 1 gene
 * Time point 120min: 0 genes


 * Time point that had the greatest number of genes significantly changed at p<0.05 was at t=60min.
 * Keeping the p-value filtered at p<0.05 and showing the genes that have an average log fold change greater than zero, we have 760 genes that meet this criteria.
 * Keeping the p-value filtered at p<0.05 and showing the genes that have an average log fold change less than zero, we have 799 genes that meet this criteria.
 * Keeping the p-value filtered at p<0.05 and showing the genes that have an average log fold change greater than 0.25, we have 727 genes that meet this criteria.
 * Keeping the p-value filtered at p<0.05 and showing the genes that have an average log fold change less than -0.25, we have 745 genes that meet this criteria.


 * The criteria that Schade et al. used to determine a significant gene expression change was that the signal intensity had to be greater than the background, it also had to be within the dynamic range of the photomultiplier tube and the raw intensities of the duplicate spots for each gene had to be within 50% within each other. Also their cut-off p-value was <0.03.  This is different than ours in the sense that we took more than two duplicates and averaged their log fold change value.  From there, we performed the T-test.  The background intensity and was taken care of when by the GenePix software.


 * NSR1:
 * Time point 15 min: average fold change: 1.20; p-value is 0.0046
 * Time point 30 min: average fold change: 1.98; p-value is 0.0180
 * Time point 60 min: average fold change: 1.96; p-value is 0.0151
 * Time point 90 min: average fold change: -0.75; p-value is 0.0676
 * Time point 120min: average fold change: -0.63; p-value is 0.1061
 * I feel that NSR1's expression is significantly chnage when we look from time point 60 min to 90 minutes.


 * The gene that has the smallest p_value is in the dtaset is gene YHL030W, at time t=90 minutes it has a p-value of 0.0000. It is a scaffold protein that assists in association of the proteasome core particle with the regulatory particle.  I think the cell is chnaging this gene's expression upon cold shock because it deals a lot with signaling pathways. The exact function of scaffold froteins is not known, but they are known to work in at least four ways, one being by regulating signal transduction by coordinating positive and negative feedback signals, which is done during cold shock.