Endocrine Abstracts (2012) 28 OC5.2

A network analysis of gene expression through childhood highlights changes related to age and growth

Adam Stevens, Andrew Whatmore & Peter Clayton


Paediatric Endocrinology, University of Manchester, Manchester, United Kingdom.


Objective: To assess age- and growth-dependent gene expression in children and correlate this with biological pathways.

Methods: We conducted a gene expression meta-analysis on datasets from normal children curated from the NCBI Gene Expression Omnibus (GEO). Four datasets were combined to form a group of 87 individuals ranging from 0.2 to 29.3 years of age (average 7.7±6.9yr). Analysis of gene expression data was performed using hierarchical clustering. Network analysis was used to determine key biological pathways involved and control points within those pathways.

Results: In the normal children there were 927 gene expression probes significantly associated with age (adjusted P-value, q<0.1) and these formed three clusters of gene expression correlating with age and the different stages of human growth: 1) <2yrs [infancy] (516 probes); 2) 10–17yrs [puberty] (242 probes) and 3) >18yrs [final height] (169 probes). Two of the genes associated with age had been previously identified as functional candidates in a meta-analysis of genetic data associated with human height1; LTBP1 (q<0.05) and IGF2BP3 (q<1×10−12). Network analysis was used to identify pathway “bottle-necks” as indicators of essential regulatory function2 and “sub-clustering” of networks demonstrated enrichment for growth and development related gene ontology. Network analysis showed that the most significant age related changes occurred in NOTCH, VEGF, TGFB and WNT pathways.

Conclusion: These data are the first observations of age related variation in the expression of genes that cluster into key growth and development pathways and therefore potential confounders when examining gene expression profiles associated with a condition or disease during childhood. 1. Lango AH, Estrada K, Lettre G et al. Hundreds of variants clustered in genomic loci and biological pathways affect human height. Nature 2010; 467(7317):832–838. 2. Yu H, Kim PM, Sprecher E, Trifonov V, Gerstein M. The importance of bottlenecks in protein networks: correlation with gene essentiality and expression dynamics. PLoS Comput Biol 2007; 3(4):e59.

Declaration of interest: There is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

Funding: No specific grant from any funding agency in the public, commercial or not-for-profit sector.

Article tools

My recent searches

No recent searches.