Example for a JProGO analysis:

Switch from aerobic to anaerobic life conditions in E. coli

Biological and experimental background:
Kang et al. (2005) performed a microarray experiment in order to identify genes that are regulated by the global transcriptional regulator FNR (Fumarate and nitrate reduction regulatory protein) in Escherichia coli. Since FNR is one key regulator for the induction and repression of genes under anaerobic conditions the authors first compared the transcriptomes of wild-type cells cultivated under aerobic and under anaerobic conditions. For both conditions, three independent cultures were grown and the RNA was prepared. Affymetrix microarrays were used for the transcriptome analysis. The complete processed data set including the expression ratios for each triplicate and the corresponding p-values (probabilities of differential expression) and q-values can be obtained as compressed Microsoft Excel file. The microarray data subset used for the analysis at hand is provided as text file .
JProGO analysis:
We analyzed the probabilities of differential expression with our JProGO tool using all implemented statistical tests. For the threshold value-based hypergeometric test and Fisher's exact test (threshold value: 0.95) no significant GO nodes were obtained (with an significance level of 0.05 and Bonferroni correction). However, for all three threshold-value independent statistical tests we found several significant nodes which are shown below both as table and as GO subgraph.
Results and Discussion:
All three threshold independent methods identified GO nodes which represent functions and process known to be affected under anaerobic conditions. For example, they recognized processes that derivate energy from the oxidation of organic compounds such as carbohydrates as significantly affected in their gene expression profile. GO nodes which reflect these processes are "main pathways of carbohydrate metabolism" and "generation of precursor metabolites and energy". In addition, the t-Test identified "succinate dehydrogenase activity" which is mainly found in the tricarboxylic acid cycle. In deed, both, Kolmogorov-Smirnov and Mann-Whitney U-test confirmed and expanded this finding since they recognized the corresponding GO node "tricarboxylic acid cycle".
Overview on performed analyses
Student's t-Test
Kolmogorov-Smirnov test
Mann-Whitney U-Test

Results of Student's t-Test:

20 significant GO nodes
back to top

1a) Result Table

GO Category GO Accession GO Name p-value
MFGO:0000104
succinate dehydrogenase activity
4.1291E-107
MFGO:0048038
quinone binding
1.7678E-11
MFGO:0005506
iron ion binding
4.0647E-9
MFGO:0016491
oxidoreductase activity
2.8895E-8
BPGO:0006091
generation of precursor metabolites and energy
1.1961E-7
MFGO:0043167
ion binding
6.3743E-7
MFGO:0046872
metal ion binding
6.3743E-7
MFGO:0016728
oxidoreductase activity, acting on CH2 groups, disulfide as acceptor
2.8317E-6
MFGO:0016151
nickel ion binding
3.69E-6
MFGO:0046914
transition metal ion binding
4.7072E-6
BPGO:0006072
glycerol-3-phosphate metabolism
6.8101E-6
MFGO:0043169
cation binding
6.9173E-6
BPGO:0006092
main pathways of carbohydrate metabolism
1.1952E-5
BPGO:0006118
electron transport
1.2248E-5
MFGO:0051539
4 iron, 4 sulfur cluster binding
2.2896E-5
BPGO:0015980
energy derivation by oxidation of organic compounds
2.4876E-5
MFGO:0051536
iron-sulfur cluster binding
2.4964E-5
MFGO:0051540
metal cluster binding
2.4964E-5
MFGO:0048037
cofactor binding
2.8086E-5
MFGO:0016651
oxidoreductase activity, acting on NADH or NADPH
4.4148E-5

1b) GO Subgraph Visualization

(Analysis Method: Student's t-test)
GO subgraph containing the nodes with a significant p-value and all their parent nodes up to the root.
Relevant GO nodes can easily be detected by their size and brightness: The lower the p-value of the GO node the larger and the brighter it is. In addition, nodes with significant p-values have a thicker border. The three sub-ontology are reflected by the color of the node (red = molecular function, green = biological process, blue = cellular component)
Click on a particalular node in order to show the genes assigned to it and their expression profile:

Results of Kolmogorov-Smirnov test:

18 significant GO nodes
back to top

2a) Result Table

GO Category GO Accession GO Name p-value
BPGO:0006092
main pathways of carbohydrate metabolism
2.3929E-12
BPGO:0006099
tricarboxylic acid cycle
4.6896E-11
BPGO:0009109
coenzyme catabolism
4.6896E-11
BPGO:0046356
acetyl-CoA catabolism
4.6896E-11
BPGO:0051187
cofactor catabolism
4.6896E-11
BPGO:0009060
aerobic respiration
1.875E-10
BPGO:0006084
acetyl-CoA metabolism
4.5661E-10
BPGO:0015980
energy derivation by oxidation of organic compounds
5.622E-10
BPGO:0006091
generation of precursor metabolites and energy
6.7633E-9
BPGO:0045333
cellular respiration
2.2937E-8
MFGO:0016491
oxidoreductase activity
9.5815E-7
BPGO:0044248
cellular catabolism
4.5524E-6
MFGO:0005506
iron ion binding
4.6734E-6
BPGO:0009056
catabolism
1.3578E-5
BPGO:0006118
electron transport
2.1169E-5
MFGO:0043167
ion binding
2.2231E-5
MFGO:0046872
metal ion binding
2.2231E-5
MFGO:0051539
4 iron, 4 sulfur cluster binding
3.6388E-5

2b) GO Subgraph Visualization

(Analysis Method: Kolmogorov-Smirnov test)
GO subgraph containing the nodes with a significant p-value and all their parent nodes up to the root.
Relevant GO nodes can easily be detected by their size and brightness: The lower the p-value of the GO node the larger and the brighter it is. In addition, nodes with significant p-values have a thicker border. The three sub-ontology are reflected by the color of the node (red = molecular function, green = biological process, blue = cellular component)
Click on a particalular node in order to show the genes assigned to it and their expression profile:

Results of Mann-Whitney U-test:

23 significant GO nodes
back to top

3a) Result Table

GO Category GO Accession GO Name p-value
BPGO:0006091
generation of precursor metabolites and energy
2.6169E-10
BPGO:0006092
main pathways of carbohydrate metabolism
9.3872E-10
BPGO:0009060
aerobic respiration
2.9706E-9
BPGO:0006099
tricarboxylic acid cycle
3.6572E-9
BPGO:0009109
coenzyme catabolism
3.6572E-9
BPGO:0046356
acetyl-CoA catabolism
3.6572E-9
BPGO:0051187
cofactor catabolism
3.6572E-9
MFGO:0016491
oxidoreductase activity
5.911E-9
MFGO:0005506
iron ion binding
6.1682E-9
BPGO:0015980
energy derivation by oxidation of organic compounds
1.4319E-8
BPGO:0006084
acetyl-CoA metabolism
1.571E-8
BPGO:0045333
cellular respiration
9.1292E-8
MFGO:0043167
ion binding
9.9237E-8
MFGO:0046872
metal ion binding
9.9237E-8
BPGO:0006118
electron transport
7.5776E-7
MFGO:0046914
transition metal ion binding
9.2834E-7
MFGO:0043169
cation binding
1.0464E-6
MFGO:0051539
4 iron, 4 sulfur cluster binding
2.5537E-6
MFGO:0051536
iron-sulfur cluster binding
1.0265E-5
MFGO:0051540
metal cluster binding
1.0265E-5
BPGO:0044248
cellular catabolism
1.292E-5
MFGO:0016151
nickel ion binding
2.5244E-5
BPGO:0009056
catabolism
2.9307E-5

3b) GO Subgraph Visualization

(Analysis Method: Mann-Whitney U-test)
GO subgraph containing the nodes with a significant p-value and all their parent nodes up to the root.
Relevant GO nodes can easily be detected by their size and brightness: The lower the p-value of the GO node the larger and the brighter it is. In addition, nodes with significant p-values have a thicker border. The three sub-ontology are reflected by the color of the node (red = molecular function, green = biological process, blue = cellular component)
Click on a particalular node in order to show the genes assigned to it and their expression profile: