Enriched Biological Functions of 1026 ROS-Diabetes Targets

Here are the enriched biological functions of 1026 ROS-Diabetes targets in terms of Gene Ontology, pathways, and protein-protein interactions. Comparison was done against the all human genes in the HUGO Gene Nomenclature Committee (HGNC) database.
Quick jump to (GO Enrichment)  (Pathway Enrichment)  (PPI Enrichment)     [ Display Reset ]



Gene Ontology Enrichment

Download a compressed file (GO_Compressed.zip (44.6 MB)) including all of the individual GO files. This zip file contains every raw and intermediate files, some of which are available on this web page.

  Corpus size # of targets (total) # of targets (used)
ROS-Diabetes set 1154 1026 1026
Background set   25254 25254

Gene Ontology Enrichment Section is composed of 9 results. For the three Gene Ontology categories (Biological Processes, Molecular Functions, and Cellular Components), three different sets of Gene Ontology terms are used.
   Explicit GO: Only use GO terms that are explicitly assigned to the target gene.
   Full GO: Use both explicitly assigned GO terms and implicitly assigned terms. Implicit terms are all the parent terms of explicity GO terms from the GO tree structure.
   Level 2~5: Use only the highest 4 levels of GO terms. These can generally provide functional categories due to their broad meanings. GO assignements are based on the full GO.

Biological Processes


Molecular Functions


Cellular Components

 


Pathway Enrichment

The following table summarizes those targets that were used in this pathway enrichment analysis. Pathway information are derived from KEGG and Reactome databases for each target.

  # of targets # of Targets with pathway info # of unique pathways
ROS-Diabetes set  1026  759  697
Background set  25254  5116  942

More full results are available.
ROS-Diabetes set (all in text file format) Pathway assignment Matrix_Type1 Matrix_Type2
Background set (all in text file format) Pathway assignment Matrix_Type1 Matrix_Type2
Back to top

Top 10 most enriched Pathways (sorted by p-value)

# of targets in ROS-Diabetes 1026 # of targets in HUGO gene set 25254 Switch table detail
Rank PathwayID Title t+ t- b+ b- p-value t-ratio b-ratio enrichment
folds
1hsa04510Focal adhesion - Homo sapiens (human)75951197250572.36e-420.070.019.4
2hsa04210Apoptosis - Homo sapiens (human)4997783251716.68e-350.050.0014.5
3hsa04010MAPK signaling pathway - Homo sapiens (human)73953259249954.29e-340.070.016.9
4109582Hemostasis62964172250824.54e-340.060.018.9
5hsa05215Prostate cancer - Homo sapiens (human)4997793251613.78e-330.050.0013.0
6hsa04920Adipocytokine signaling pathway - Homo sapiens (human)4598172251825.99e-330.040.0015.4
7hsa04910Insulin signaling pathway - Homo sapiens (human)53973137251171.68e-300.050.019.5
8hsa05212Pancreatic cancer - Homo sapiens (human)4298474251801.42e-290.040.0014.0
9168256Signaling in Immune system72954312249425.52e-290.070.015.7
10hsa05222Small cell lung cancer - Homo sapiens (human)4398387251672.28e-280.040.0012.2

Full result (HTML,   TXT,   EXCEL)   Back to top 

 


Protein-Protein Interaction Network

The following table summarizes the targets (from the ROS-Diabetes set) that were used in this PPI network analysis. Background set is the source of the target frequency information for random network generation. For more detail, refer to the Network Significance Test below.

  Corpus size # of targets (total) # of targets (used)
ROS-Diabetes 1154 1026 1026
Background set   25254 25254

Interaction Network Within Corpus (sorted by degree)

[TABLE] Here is a summary table of the Protein-Protein Interaction network data of the test set. This table basically shows how many targets are interacting each other within the test set.

Num of PMIDs 1154
Num of genes (total) 1026
Num of genes (used) 1026
Num of targets with any interaction 983
 Num of targets interacting within corpus  879
 Num of direct interaction within corpus  5002
Maximum degree target HNF4A   (Degree = 173)
Minimum degree target SLC7A2   (Degree = 1)

[TABLE] Top 10 interacting targets within the test set

 Rank  Target Symbol  Num of Direct Interactions  MiMI Network
 1  HNF4A  173 
 2  JUN  106 
 3  MYC  102 
 4  TP53  82 
 5  MAPK1  77 
 6  SRC  75 
 7  RELA  74 
 8  PRKCA  72 
 9  EGFR  70 
 10  MAPK3  62 

View your full list in a TEXT or EXCEL file.
View the full merged network in cytoscape SIF (MiMI Network File).
This file has to be downloaded first and imported into a Cytoscape program.

Network Significance Test (100 iteration of random genes with size of 1026 from total of 25254 genes)

This Network Significance Test section will test the integrity of the Protein-Protein Interaction network among the test set targets. This is based on the assumption that targets commonly related to a certain topic will be more likely to have frequent protein-protein interaction with each others. Therefore, if a set of targets identified by text-mining (SciMiner) from a certain query have more frequent direct interactions among the targets compared to randomly generated sets, it can, in part, support the validity of using text-mining method to identify related targets from a set of related papers. 100 random network of the same number of target in the test set will be generated from the background set. The test set had 1026 targets. Thus 1026 are randomly selected from the background which had a total of 25254 targets. If the background set was full HUGO set, all the targets will have an equal chance of being selected. If the background set was either Selected background above or Whole document in SciMinerDB, the probability of each target begin selected is determined by the observed frequency of each target as in (# of papers with the target / # of all papers).Two statistical measures are given below; standard Z-test and one-sample T-test.

  Targets with any interaction Targets interacting within corpus Direct interactions within corpus Max degree
Tested 983 879 5002 173
Mean 529.75 158.59 171.41 25.10
STDEV 14.76 34.00 51.47 35.93
Z-Score 30.7 21.2 93.9 4.1
P-value(Z) 0.0e+00 (0.000) 0.0e+00 (0.000) 0.0e+00 (0.000) 1.9e-05 (0.000)
T-Stat -307.0 -211.9 -938.5 -41.2
P-value(T) 2.7e-149 (0.000) 2.3e-133 (0.000) 2.6e-197 (0.000) 4.4e-64 (0.000)

Distrubution of randomly generated networks

This section will histograms of the random network samples. Red bars represent the targets from the ROS-Diabetes set (target query) being tested.
Targets with any interaction Targets interacting within corpus Direct interactions within corpus Max degree
 
Back to top 

 

SciMiner Homepage