Limits...
Functional networking of human divergently paired genes (DPGs).

Xie B, Wang D, Duan Y, Yu J, Lei H - PLoS ONE (2013)

Bottom Line: Here we report a network-based comprehensive analysis on human DPGs and our results indicate that the two members of the DPGs tend to participate in different biological processes while enforcing related functions as modules.The functional network bridged by DPGs consists of three major modules.This module also shows low variation in expression in both CNS (central nervous system) and non-CNS tissues.

View Article: PubMed Central - PubMed

Affiliation: CAS Key Laboratory of Genome Sciences and Information, Beijing Institute of Genomics, Chinese Academy of Sciences, Beijing, China ; University of Chinese Academy of Sciences, Beijing, China.

ABSTRACT
Divergently paired genes (DPGs), also known as bidirectional (head-to-head positioned) genes, are conserved across species and lineages, and thus deemed to be exceptional in genomic organization and functional regulation. Despite previous investigations on the features of their conservation and gene organization, the functional relationship among DPGs in a given species and lineage has not been thoroughly clarified. Here we report a network-based comprehensive analysis on human DPGs and our results indicate that the two members of the DPGs tend to participate in different biological processes while enforcing related functions as modules. Comparing to randomly paired genes as a control, the DPG pairs have a tendency to be clustered in similar "cellular components" and involved in similar "molecular functions". The functional network bridged by DPGs consists of three major modules. The largest module includes many house-keeping genes involved in core cellular activities. This module also shows low variation in expression in both CNS (central nervous system) and non-CNS tissues. Based on analyses of disease transcriptome data, we further suggest that this particular module may play crucial roles in HIV infection and its disease mechanism.

Show MeSH

Related in: MedlinePlus

Comparison of GO similarity between DPGs and random gene pairs.Frequence distributions of GO similarity score on biological process (BP, A), cell component (CC, B), and molecular function (MF, C). The blue and red lines depict frequency distributions of DPGs and the randomly sampled gene pairs, respectively.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3815023&req=5

pone-0078896-g001: Comparison of GO similarity between DPGs and random gene pairs.Frequence distributions of GO similarity score on biological process (BP, A), cell component (CC, B), and molecular function (MF, C). The blue and red lines depict frequency distributions of DPGs and the randomly sampled gene pairs, respectively.

Mentions: As physically neighbouring pairs, the functional similarity within a DPG is of great interest. We calculated the GO similarity score of gene pairs of DPGs (details in Methods) and compared the distribution of these scores to random pairing using Kolmogorov-Simirnov test. Our main findings are four folds. First, DPGs are not significantly different from random gene pairs in biological process (BP; p value = 0.06234) but significantly different in cell component (CC; p value = 7.44E-15) and molecular function (MF; p value  = 1.73E-11). In details, we observed the left shift of the distribution line of BP and the right shift of those of CC and MF, in comparison to the background (BG; Figure 1). Second, gene pairs with paralogs in DPGs somewhat influence the result of this similarity analysis. We can see a peak close to score 1.0 on the map of BP but the DPG pairs attributed to this peak are largely 12 pairs of histone genes. If we exclude these histone pairs and then compare to the background, the p value becomes 0.2559. We also found that these 12 gene pairs form nearly independent modules in DPG functional networking. Third, DPGs have higher tendency to be paired for similar cellular components. In addition to the smaller p value and the rightward shift of the distribution, we also found that the right peak is much higher than the left one on the map of CC, contrary to the background. Fourth, there is no significant correlation between TSS distance of DPGs and functional similarity, regardless of BP, CC, or MF.


Functional networking of human divergently paired genes (DPGs).

Xie B, Wang D, Duan Y, Yu J, Lei H - PLoS ONE (2013)

Comparison of GO similarity between DPGs and random gene pairs.Frequence distributions of GO similarity score on biological process (BP, A), cell component (CC, B), and molecular function (MF, C). The blue and red lines depict frequency distributions of DPGs and the randomly sampled gene pairs, respectively.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3815023&req=5

pone-0078896-g001: Comparison of GO similarity between DPGs and random gene pairs.Frequence distributions of GO similarity score on biological process (BP, A), cell component (CC, B), and molecular function (MF, C). The blue and red lines depict frequency distributions of DPGs and the randomly sampled gene pairs, respectively.
Mentions: As physically neighbouring pairs, the functional similarity within a DPG is of great interest. We calculated the GO similarity score of gene pairs of DPGs (details in Methods) and compared the distribution of these scores to random pairing using Kolmogorov-Simirnov test. Our main findings are four folds. First, DPGs are not significantly different from random gene pairs in biological process (BP; p value = 0.06234) but significantly different in cell component (CC; p value = 7.44E-15) and molecular function (MF; p value  = 1.73E-11). In details, we observed the left shift of the distribution line of BP and the right shift of those of CC and MF, in comparison to the background (BG; Figure 1). Second, gene pairs with paralogs in DPGs somewhat influence the result of this similarity analysis. We can see a peak close to score 1.0 on the map of BP but the DPG pairs attributed to this peak are largely 12 pairs of histone genes. If we exclude these histone pairs and then compare to the background, the p value becomes 0.2559. We also found that these 12 gene pairs form nearly independent modules in DPG functional networking. Third, DPGs have higher tendency to be paired for similar cellular components. In addition to the smaller p value and the rightward shift of the distribution, we also found that the right peak is much higher than the left one on the map of CC, contrary to the background. Fourth, there is no significant correlation between TSS distance of DPGs and functional similarity, regardless of BP, CC, or MF.

Bottom Line: Here we report a network-based comprehensive analysis on human DPGs and our results indicate that the two members of the DPGs tend to participate in different biological processes while enforcing related functions as modules.The functional network bridged by DPGs consists of three major modules.This module also shows low variation in expression in both CNS (central nervous system) and non-CNS tissues.

View Article: PubMed Central - PubMed

Affiliation: CAS Key Laboratory of Genome Sciences and Information, Beijing Institute of Genomics, Chinese Academy of Sciences, Beijing, China ; University of Chinese Academy of Sciences, Beijing, China.

ABSTRACT
Divergently paired genes (DPGs), also known as bidirectional (head-to-head positioned) genes, are conserved across species and lineages, and thus deemed to be exceptional in genomic organization and functional regulation. Despite previous investigations on the features of their conservation and gene organization, the functional relationship among DPGs in a given species and lineage has not been thoroughly clarified. Here we report a network-based comprehensive analysis on human DPGs and our results indicate that the two members of the DPGs tend to participate in different biological processes while enforcing related functions as modules. Comparing to randomly paired genes as a control, the DPG pairs have a tendency to be clustered in similar "cellular components" and involved in similar "molecular functions". The functional network bridged by DPGs consists of three major modules. The largest module includes many house-keeping genes involved in core cellular activities. This module also shows low variation in expression in both CNS (central nervous system) and non-CNS tissues. Based on analyses of disease transcriptome data, we further suggest that this particular module may play crucial roles in HIV infection and its disease mechanism.

Show MeSH
Related in: MedlinePlus