SAM - Analysis tree

SAN analysis generates Delta values and lists of statistically significant and non-significant genes.Tto view these data you will find a series of nodes in the the Analysis tree:

Parameter:

View parameters used for this particular SAM analysis:

Click the Copy button to copy the displayed parameters into Clipboard.

Delta table

Show  numbers of significant / non-significant genes depending on Delta value:

SUMO calculates for each Delta value between 0 and 10 with increment 0.01 SAM statistics and summarizes the results in the Delta table:

• Delta = Delta value
• Cut-high :  D_experimental where D_experimental - D_permutation > Delta
• Cut-low :  D_experimental where D_permutation -   D_experimental > Delta
• + significant = number of up-regulated significant genes
• + false = number of false significant up-regulated genes
• + FDR = false discovery rate for up-regulated genes
• - significant = number of down-regulated significant genes
• - false = number of false significant down-regulated genes
• - FDR = false discovery rate for down-regulated genes
• significant = number of regulated significant genes (= "+ significant" + "- significant")
• + false = number of false significant regulated genes
• + FDR = false discovery rate for regulated genes

To save the table selec File | Save as. For more details how to work with data tables go here.

Delta graph

These graphs visualizes the values from the Delta table as line graphs.
They can be used to roughly estimate a sufficient Delta value.

Delta graph absolute

This graph show abolute numbers of genes (y-axis) in dependance of the Delta value (x-axis).

• Red line:  significant = number of up-regulated significant genes
• Brown line:  + false = number of false significant up-regulated genes
• Green line:  significant = number of down-regulated significant genes
• Dark green line: - false = number of false significant down-regulated genes
• Blue line:  significant = number of regulated significant genes (= "+ significant" + "- significant")
• Purple line: non-significant = number of false significant regulated genes

Delta graph relative

This graph shows FDR (y-axis) in dependance of the Delta value (x-axis).

For more details how to work with data tables go here.

SAM graph

The SAM graph visualizes the distribution of experimental statistics ( D_exp ) versus permutaion statistics  (D_perm ).
Therfore, both datasets are ranked (i.e. sorted) and siplayed as scatterplot.

Elements of the graph:

• Main diagon (black line)
• Delta threshold lines (red for up-regulated genes, green for downregulated genes).
• Blue squares represent all individual genes

Controls

• Delta slider: use the slider to adjust Delta value. Present value is shown in Delta value edit field
• Delta edit field: type a nubmer for teh Delt value

Data table

show number for significant and non significant gens as well as FDRs.

Use the Delta slider to adjust the delta value such, that a desired FDR or number of singnificant / non-significant genes can be filtered:

Example:

• Left panel: Delta set to have = = 0% FDR
• Right panel: Delta set to have ~ 5% FDR

Genelists, Gene expression sub-matrices, Gene profiles, Controid profiles and Heatmaps show Genes filtered according to the present Delta settings.

Gene Lists

Get lists of significant / non-significant genes:

Click a subnode to view the respective list of genes:

• positive significant genes= significant up-regulated genes
• negative significant genes= significant down-regulated genes (only in single- and two-class SAM)
• all significant genes = positive + negative (only in single- and two-class SAM)
• NON significant genes

The table shows:

• Gene # = Original gene number is source expression matrix
• All gene annotations as found in the original source expression matrix
• Di-Experimental
• Di-Permutation

For more details how to work with data tables go here.

Gene expression sub-matrices

Get expression sub-matrix of respective genes with all expression values for subsequent analysis.

Click a subnode to view the respective sub-matrix :

• positive significant genes= significant up-regulated genes
• negative significant genes= significant down-regulated genes (only in single- and two-class SAM)
• all significant genes = positive + negative (only in single- and two-class SAM)
• NON significant genes

Gene profiles

View expression profiles of the respective genes as line graphs:

Click a subnode to view the respective gene's expression profiles :

• positive significant genes= significant up-regulated genes
• negative significant genes= significant down-regulated genes (only in single- and two-class SAM)
• all significant genes = positive + negative (only in single- and two-class SAM)
• NON significant genes

Example:

• Left panel: up-regulated significant genes
• Right panel: down-regulated significant genes

When thousands of genes are in a selection group (e.g. NON significant genes), it may take a while to build the graph.
For more details how to work with data tables go here.

Centroid profiles

View averaged expression profile from all respective genes:

Click a subnode to view the averaged expression profiles from all respective genes :

• positive significant genes= significant up-regulated genes
• negative significant genes= significant down-regulated genes (only in single- and two-class SAM)
• all significant genes = positive + negative (only in single- and two-class SAM)
• NON significant genes

Example:

• Left panel: up-regulated significant genes
• Right panel: down-regulated significant genes

Heat maps

View expression profiles of the respective genes as Heatmaps:

Click a subnode to view the respective sub-matrix as hetmap:

• positive significant genes= significant up-regulated genes
• negative significant genes= significant down-regulated genes (only in single- and two-class SAM)
• all significant genes = positive + negative (only in single- and two-class SAM)
• NON significant genes

Example:

• Left panel: up-regulated significant genes
• Right panel: down-regulated significant genes

last edited 04.04.2006