Personal tools

Coexpression cluster:C4100

From FANTOM5_SSTAR

Revision as of 15:40, 19 October 2012 by Autoedit (talk | contribs)
Jump to: navigation, search


Full id: C4100_chronic_Basophils_acute_Natural_Whole_CD34_leukemia



Phase1 CAGE Peaks

Hg19::chr22:38539430..38539442,-p10@PLA2G6
Hg19::chr22:38539452..38539464,-p7@PLA2G6
Hg19::chr22:38539487..38539506,-p6@PLA2G6


Enriched pathways on this co-expression cluster<b>Summary:</b><br>Canonical pathway gene sets were compiled from Reactome, Wikipathways and KEGG. For the major signaling pathways, the transcriptionally-regulated genes (downstream targets) were obtained from Netpath. Combined, the canonical pathways and downstream targets totaled 489 human gene sets. The corresponding M. musculus gene sets were inferred by homology using the HomoloGene database. Enrichment for each of the canonical 489 pathways and gene sets included in the co-expression cluster was assessed by the hypergeometric probability. The resulting P values were also then adjusted by the Benjamini-Hochberg method for multiple comparisons.<br><b>Analyst: </b>Emmanuel Dimont<br><br>link to source dataset<br>data


No results for this coexpression

Enriched Gene Ontology terms on this co-expression cluster<b>Summary:</b> Results for GOStat analysis on co-expressed clusters. Each cluster with promoters mapping to at least two different genes was analysed with GOStat (PMID: 14962934) with default parameter. <br><b>Analyst:</b> Erik Arner<br><br>link to source dataset<br>data


No GOStat results

Enriched sample ontology terms on this co-expression cluster<b>Summary:</b>To summarize promoter activities (expression profile of a TSS region) across ~1000 samples, we performed enrichment analysis based on FANTOM5 Sample Ontology (FF ontology). The question here is “in which type of samples the promoter is more active”. To answer this question, we compared expressions (TPMs) in the samples associated with a sample ontology term and the rest of the samples by using the Mann-Whitney rank sum test. To summarize ontologies enriched in this co-expression cluster, we ran the same analysis on an averaged expression profile of all promoters that make up. <b>Analyst:</b> Hideya Kawaji <br><br>links to source dataset<br><br>cell_data<br>uberon_data<br>disease_data<br>


Uber Anatomy
Ontology termp-valuen
blood1.22e-1715
haemolymphatic fluid1.22e-1715
organism substance1.22e-1715
Disease
Ontology termp-valuen
myeloid leukemia3.61e-1831
leukemia2.62e-1639
hematologic cancer8.46e-1251
immune system cancer8.46e-1251
chronic leukemia6.32e-098


Overrepresented TFBS (DNA) motifs on this co-expression cluster<b>Summary:</b>The values shown are the p-values for overrepresentation of the motif in this coexpression cluster. So a small p-value means a strong overrepresentation. <b>Analyst:</b> Michiel de Hoon <br><br>link to source data <br> Novel motifs <br>data <br><br> Jaspar motifs <br>data


Novel motifs



JASPAR motifs

Motifs-log10(p-value)

{{{tfbs_overrepresentation_jaspar}}}



ENCODE TF ChIP-seq peak enrichment analysis<b>Summary:</b> For each TF and each co-expression cluster, the number of promoters with ENCODE TF ChIP signal was compared with the rest of promoters from the robust set using Fisher's exact test. Clusters with significant ChIP enrichment (q <= 0.05) after Benjamini-Hochberg correction were retained. <br><b>Analyst:</b> Erik Arner<br><br>link to source dataset<br><br>data


(#promoters = Number of promoters in this coexpression cluster that have ChIP signal of the TF)

TF#promotersEnrichmentp-valueq-value
CCNT2#90536.336201576962630.003930750035764890.0190383343219983
E2F4#1874312.66806031528440.0004917987006298980.00437839608687119
ELF1#199734.258097958807540.01295179875054610.0463275878384918
GABPB1#255337.067683836182170.002832212825417420.0154396844803095
GATA1#2623313.56030814380040.0004009615963782630.00388388315378867
GATA2#2624312.7449317335540.0004829527704283790.00437383839670218
HDAC2#3066313.41562023662630.0004140761399857210.00391966808519611
TAL1#6886329.86861667744023.75103522793067e-050.000720534059536235
TFAP2A#7020316.5186343730450.0002218033880766340.00248630718550344
TFAP2C#7022310.80922860986020.0007916746575753130.00616751525902641



Relative expression of the co-expression cluster<b>Summary:</b>Co-expression clusters are compared against FANTOM5 samples to obtain relative expression. <br><b>Analyst:</b>NA<br><br>link to data source<br> data


This analysis result is provided for C0 - C305 clusters.