JPBS

Background: Recently, HSPA1A and YBX3 have been suggested as influential genes for Autism Spectrum Disorder (ASD). However, the results have shown a lack of replication.Method: Here we performed a systematic review and a meta-analysis to explore the gene expression activity of HSPA1A and YBX3 in ASD. Fifteen RNA expression datasets of ASD case vs. control studies were acquired from Gene Expression Omnibus, including 1,534 samples (830 ASD cases, and 704 healthy controls). Both fixed-effect and random-effects models were employed for meta-analysis. Multiple linear regression (MLR) were employed to study the possible influence of the following factors: tissue, sample size, sex and racial region. Results: Utilizing all 15 datasets, both genes presented no significant log fold change (LFC) in ASD: YBX3 (LFC = 0.001; p-value = 0.49); HSPA1A (LFC = 0.115; p-value = 0.16), but weak effect sizes for the 3 brain tissue sampled datasets: YBX3 (LOR = 0.89; p-value = 0.11); HSPA1A (LOR = 0.91; p-value = 0.086). Sample tissue is a major factor for both genes (p-value < 1.5e-04) and sex is for HSPA1A only (p-value < 0.01). Literature based pathway analysis show that YBX3 and HSPA1A may play inhibitory roles again ASD. Conclusion: Our results suggest that YBX3 and HSPA1A may present weak protective regulation within the brain of ASD patients.

Autism Spectrum Disorder; RNA expression data; Random model; Multiple linear regression model