From The Polyploidy Portal
Contents |
Figures
figure 1
Figure 1. Comparing gene expression using microarrays. mRNA is extracted from a plant that has undergone an experimental treatment (T, drought stress in this case) and an untreated control (C). The mRNA undergoes reverse transcription to generate a cDNAs. A different fluorescent molecule is used to label each of the cDNA pools. The labeled cDNAs are then hybridized to a microarray. The microarray consists of a glass slide on which thousands of distinct DNA sequences have been affixed. Each dot (or “feature”) on the slide represents the sequence of a different plant gene. After the unbound probe is washed away, a special slide scanner excites each feature on the array with a laser and measures the fluorescent signal emitted. The more cDNA is bound to a spot, the greater the signal will be. The magnified computer screen at the lower right shows the possible results for each feature. A red spot (A) represents a gene that is only expressed in the control. A green spot (B) represents a gene that is only expressed in the treated plant. A yellow spot (C) represents a gene that is expressed in both treated and control plants. A dark spot (D) indicates that the corresponding gene is not expressed in either the control or treated plants.
figure 2
Figure 2. Sources of variation in gene expression studies.
figure 3
Figure 3. Spot intensity. Aftter labeled cDNA is hybridized to a microarray each fluorescent label is individually excited and detected by the slide scanner. The scanner divides the spots into pixels. The spot are not uniform and there can be variable intensity from pixel to pixel within each spot. Also, there are areas of low intensity, “background” fluorescence in the regions between spots. After scanning, the red and green signals are superimposed by the computer that generates a composite yellow spot (notice the distinct red and green pixels in the background of the superimposed image.)
figure 5
Figure 5. Using an MA plot to visualize normalization on one array. Highly up- or down-regulated genes are above or below the x-axis.. The diagonal lines to the left are called “fishtails”. Fishtails are an artifact of changing the value for spots that have higher background than foreground.
figure 6
Figure 6: Experimental design for a dye-swap experiment. In this design, the treatment and control are first labeled with one set of dyes and hybridized to array 1. To account for different labeling efficiencies of the two dyes, the same probes are now labeled with the other dye (the dyes are swapped) and subsequently hybridized to array 2.
figure 7
Figure 7. Is the average M equal to zero? Multiple repeats for the same gene will give different results. Statistical tests provide an answer whether or not the mean of the repeats is significantly different from zero, or, in other words, if the treatment resulted in differences in gene expression from the control. On the left different means for observations with the same variation pattern are shown. The variation in the measurements is important in deciding whether the mean of a number of observations is significantly different from zero. If the variation is small, we may be more inclined to assume a non-zero mean than if the variation is large.
figure 8
Figure 8. t-distribution for different degrees of freedom.