The results of any query are displayed in table form. Each row in the table contains links to all of the information available for a given gene. Each column provides a link to a specific kind of data or information. The column headings describe the kind of information that the link (a colored button or a icon) goes to. The purple sidebar on the left side of each table contains links to information pertinant to the gene or entire list of genes in the table.
ORF, Gene Name
The ORF is the systematic name (or ORF name) of the gene and reflects the chromosomal position of the gene. The Gene Name is the assigned standard name (or standard locus name).
Change in dataset
- Max
Refers to the Maximum increase in expression observed for that gene in the entire dataset. A minus sign in the Max-Increase column indicates the expressions in all experiment conditions for the given gene are decreased.
- Max
Refers to the Maximum decrease in expression observed for that gene in the entire dataset. A minus sign in Max-Decrease column indicates that the expressions in all experiment conditions for the specified gene are increased.
- vs. other datasets Refers to the Histogram
Back to top
Histogram (vs. Other Datasets)
The histogram helps users see the observed range of gene expression
changes in every experiment in every dataset in the database. The
data are in log2 space meaning that positive numbers represent
increases in expression and negative numbers represent decreases in
expression. For instance, a log2 ratio of 1.0 indicates an increase
of the transcript level by 2-fold and -1.0 indicates a decrease by
2-fold. A log2 ratio of zero would indicate no change. The log2 ratios
are rounded to the nearest 0.1 value for this display. To see the
actual experiments and datasets where these changes occur, click on
individual bars in the bar graph.
Thumbnail Display
Thumbnail Display is a miniature version of the 3-color graphical expression display. To see a large version with labels, click on the "Enlarge" link in the column header.
The page with the 'enlarged' graphical expression displays also shows the GO annotations associated with each gene. When more than one annotation exists, an asterisk is shown. You can see all of the GO terms for a gene by going to the locus page via the hyper-linked ORF name.
If you would like to analyze the list of genes further (Do they share a common function, act in a common process, or share a cellular component?), follow the links to the GO Term Mapper and GO Term Finder. See the "Help" button at the top of each page for more info, or the GO Resources page.
Back to top
Graphical Expression Display
Expression data is often displayed with a 3-color graphic: 
Each square of color represents the expression level of a gene in a single experiment in the dataset. A miniature version is given for each gene in the thumbnail display column on the results pages. A larger version can be found by clicking the "Enlarge" link.
The convention followed by the vast majority of microarray experiments is this:
Red means the mRNA level (expression) is increased, relative to the standard.
Black means the mRNA level (expression) is unchanged, relative to the standard.
Green means the mRNA level (expression) is decreased, relative to the standard.
This scale is used as a reference to help determine the relative expression levels within the 3-color representation of the data (graphical expression display).
Note that there are exceptions to this general rule. For example, in the data set entitled "Gene regulation by HTZ1, SWR and SIR2",
because of a dye quenching artifact observed with abundant mRNAs, the authors perform each experiment in quadruplicate and reverse
the fluors. As a result the displayed colors alternate between red and green from replicate-to-replicate but the trends are still the
same so that similarly expressed genes can be identified.
Similarly Expressed Genes
This link goes to a page much like the results display in the original Expression Connection. Up to 20 genes expressed in a similar pattern to the gene of interest are listed along with the 3-color graphical expression display and the GO terms associated with each gene. Two genes are considered to be similarly expressed if the Pearson correlation is above 0.8. O.8 is an arbitrary cut-off, but one that is considered rather stringent.
If you would like to analyze the list of genes further (Do they share a common function, act in a common process, or share a cellular component?), follow the links to the GO Term Mapper and GO Term Finder. See the "Help" button at the top of each page for more info, or the GO Resources page.
Back to top
Expression Graph
The Expression Graph is an alternative to the 3-color graphical expression display for displaying data. The 3-color graphical expresssion display often does not offer enough differentiation between data points, especially at the spectrum's extremes. In addition, the timecourse or concentration course of the dataset are not obvious. On the Expression Graph, the numerical values for expression changes are plotted versus a scaled x-axis (time, concentration and etc.), giving the user a better feel for the amplitutude and course of expression changes in the dataset.
Pearson Correlation
The Pearson Correlation is a graphical representation of the Pearson correlation coefficients for each pair of genes in a gene list. A Pearson correlation coefficient is a statistical measure of the extent of correlation between two independent items. In this setting it reflects the extant of correlation between the expression patterns of two genes in a microarray dataset.
The color of each square indicates the extent of correlation between the expression patterns of two genes. A Pearson correlation of 1.0 (black in the graphic) indicates perfect correlation, a calue of 0 indicated no correlation, and a score of -1 indicates perfect anti-correlation. Thus the closer to 1 the Pearson correlation is, the better correlated two profiles are and the darker red the square is. Each gene calculated against itself is perfectly correlated, hence the diagonal has all black squares. SGD uses 0.8 (an arbitrary cut-off) to indicate which gene pairs have expression patterns that are well correlated (dark red in the graphic).
The Pearson coefficient (phi G) is calculated by:
| Equation 1 |
|
| Equation 2 |
|
In equation 1 S(X,Y) is identical to the textbook Pearson Correlation if Goffset in equation 2 is set to the mean of the observations. For the calculations in Expression Connection, the Goffset is set to 0 (zero), calculating the Pearson correlation of uncentered metrics.
The program that calculate the Pearson Correlation was provided kindly by Gavin Sherlock, Stanford Microarray Database (SMD). See SMD help documentation for further information.
Back to top
GO Term Mapper
The GO Term Mapper takes a list of genes and maps the granular GO annotations assigned by SGD curators to more general terms, chosen by the user. This allows users to ask whether a group of genes share particular commonalities (function, process, cellular component). For more help, click the help button at the top of the GO Term Mapper page.
GO Term Finder
The GO Term Finder takes a list of genes and maps the granular GO annotations assigned by SGD curators to more general significant terms. This allows users to find any commonalities (in function, process, and cellular component) that might exist among genes. For more help, click the help button at the top of the GO Term Finder page.
Stats
The green Stats icon links to a summary table containing information on how many genes in the dataset were increased, decreased, and etc. in each experiment in the dataset. SGD downloads expresssion data from websites accompanying the publication of the data. These statistics are calculated from the data and are not given to SGD by authors.
The title of the dataset is displayed at the top of the summary statistics page. The statistics for individual experiments is contained in the rows of the table. The columns contain the statistics. All numbers are expressed as a percent of the total spots (distinct mRNAs) in the experiment. The percent Increased or Decreased refers to the percent of genes that change expression at least 2-fold relative to the control.
Regulators
A gene can be a Regulator of other genes and be Regulated
by other genes. Eran Segal and colleagues have evaluated the
expression patterns of genes in certain microarray datasets to
elucidate potential regulatory modules, which consist of genes
designated as regulators
and regulatees, and possible regulatory motifs found in intergenic
regions that may mediate the regulation. Eran Segal and colleagues have provided the web pages with the results of their analyses to SGD. For more help, see their comprehensive website.
Red squares in the search menu link to the search forms below. Red squares next to each search form link bact to the search menu at the top of the page.
Return to Saccharomyces Genome Database