We used conventional statistics and artificial neural networks (ANNs) to unravel connections among BDNF, OXTR, HTR1A, RELN, EN2, BCL2, and MECP2 gene methylation levels, sex, maternal risk factors and symptoms severity (ADOS-2 score) in a cohort of 58 children with autism spectrum disorders (ASD). Sex differences were observed in blood DNA methylation levels of the studied genes, and ANNs revealed sex-specific connections among maternal risk factors and gene methylation. Furthermore, ANNs selected a set of variables allowing discriminating between high and low-moderate ADOS-2 scores with 86.8% overall accuracy. Particularly, high gestational weight gain, lack of folic acid supplements, advanced maternal age, pre-term birth, low birthweight, and living in rural context were the best predictors of high ADOS-2 score. Moreover, the analysis of saliva DNA samples revealed that Mir-28 methylation levels could represent a biomarker of disease severity in ASD children. In conclusion, ANNs revealed links among ASD maternal risk factors, symptoms severity and gene methylation levels, as well as sex differences in gene methylation levels that warrant further investigation in ASD.

Notes:

1-Dept of Translational Research and of New Surgical and Medical Technologies, University of Pisa, Pisa, Italy
2-IRCCS Stella Maris Foundation, Calambrone, Italy
3-Villa Santa Maria Foundation, Tavernerio, Italy