Non-biological experimental variation or "batch effects" are commonly observed across multiple batches of microarray experiments, often rendering the task of combining data from these batches difficult. The ability to combine microarray data sets is advantageous to researchers to increase statistical power to detect biological phenomena from studies where logistical considerations restrict sample size or in studies that require the sequential hybridization of arrays. In general, it is inappropriate to combine data sets without adjusting for batch effects. Methods have been proposed to filter batch effects from data, but these are often complicated and require large batch sizes ( > 25) to implement. Because the majority of microarray studies are conducted using much smaller sample sizes, existing methods are not sufficient. We propose parametric and non-parametric empirical Bayes frameworks for adjusting data for batch effects that is robust to outliers in small sample sizes and performs comparable to existing methods for large samples. We illustrate our methods using two example data sets and show that our methods are justifiable, easy to apply, and useful in practice. Software for our method is freely available at: http://biosun1.harvard.edu/complab/batch/.

We previously observed that nitric oxide (NO) exposure increases the stability of mRNAs encoding heme oxygenase 1 (HO-1) and TIEG-1 in human and mouse fibroblasts. Here, we have used microarrays to look broadly for changes in mRNA stability in response to NO treatment. Using human IMR-90 and mouse NIH 3T3 fibroblasts treated with actinomycin D to block de novo transcription, microarray analysis suggested that the stability of the majority of mRNAs was unaffected. Among the mRNAs that were stabilized by NO treatment, seven transcripts were found in both IMR-90 and NIH 3T3 cells (CHIC2, GADD45B, HO-1, PTGS2, RGS2, TIEG, and ID3) and were chosen for further analysis. All seven mRNAs showed at least one hit of a signature motif for the stabilizing RNA-binding protein (RBP) HuR; accordingly, ribonucleoprotein immunoprecipitation analysis revealed that all seven mRNAs associated with HuR. In keeping with a functional role of HuR in the response to NO, a measurable fraction of HuR increased in the cytoplasm following NO treatment. However, among the seven transcripts, only HO-1 mRNA showed a robust increase in the level of its association with HuR following NO treatment. In turn, HO-1 mRNA and protein levels were significantly reduced when HuR levels were silenced in IMR-90 cells, and they were elevated when HuR was overexpressed. In sum, our results indicate that NO stabilizes mRNA subsets in fibroblasts, identify HuR as an RBP implicated in the NO response, reveal that HuR alone is insufficient for stabilizing several mRNAs by NO, and show that HO-1 induction by NO is regulated by HuR.