ADAGE signature analysis: differential expression analysis with data-defined gene sets.

BACKGROUND: Gene set enrichment analysis and overrepresentation analyses are commonly used methods to determine the biological processes affected by a differential expression experiment. This approach requires biologically relevant gene sets, which are currently curated manually, limiting their availability and accuracy in many organisms without extensively curated resources. New feature learning approaches can now be paired with existing data collections to directly extract functional gene sets from big data. RESULTS: Here we introduce a method to identify perturbed processes. In contrast with methods that use curated gene sets, this approach uses signatures extracted from public expression data. We first extract expression signatures from public data using ADAGE, a neural network-based feature extraction approach. We next identify signatures that are differentially active under a given treatment. Our results demonstrate that these signatures represent biological processes that are perturbed by the experiment. Because these signatures are directly learned from data without supervision, they can identify uncurated or novel biological processes. We implemented ADAGE signature analysis for the bacterial pathogen Pseudomonas aeruginosa. For the convenience of different user groups, we implemented both an R package (ADAGEpath) and a web server ( http://adage.greenelab.com ) to run these analyses. Both are open-source to allow easy expansion to other organisms or signature generation methods. We applied ADAGE signature analysis to an example dataset in which wild-type and ∆anr mutant cells were grown as biofilms on the Cystic Fibrosis genotype bronchial epithelial cells. We mapped active signatures in the dataset to KEGG pathways and compared with pathways identified using GSEA. The two approaches generally return consistent results; however, ADAGE signature analysis also identified a signature that revealed the molecularly supported link between the MexT regulon and Anr. CONCLUSIONS: We designed ADAGE signature analysis to perform gene set analysis using data-defined functional gene signatures. This approach addresses an important gap for biologists studying non-traditional model organisms and those without extensive curated resources available. We built both an R package and web server to provide ADAGE signature analysis to the community.

Polymorphisms in the PON gene cluster are associated with Alzheimer disease.

Paraoxonase is an arylesterase enzyme that is expressed in the liver and found in the circulation in association with apoA1 and the high-density lipoprotein, and prevents the accumulation of oxidized lipids in low-density lipoproteins in vitro. Common polymorphisms in genes encoding paraoxonase are established risk factors in a variety of vascular disorders including coronary artery disease and carotid artery stenosis, but their association with Alzheimer disease (AD) is controversial. We tested the association of 29 SNPs in PON1, PON2 and PON3 with AD in 730 Caucasian and 467 African American participants of the MIRAGE Study, an ongoing multi-center family-based genetic epidemiology study of AD. Eight SNPs were associated with AD in the African American families (0.0001< or =P< or =0.04) and two SNPs were associated with AD in Caucasian families (0.01< or =P< or =0.04). Of note, the pattern of association for the PON1 promoter SNP -161[C/T] was the same in both ethnic groups (P=0.006). Haplotype analysis using sliding windows revealed 11 contiguous SNP combinations spanning the three PON genes with significant global test scores (0.006< or =P< or =0.04) in the two ethnic groups combined. The most significantly associated haplotype comprised SNPs in the region spanning the -161[C/T] SNP (P=0.00009). Our results demonstrate association between AD and variants in the PON gene cluster in Caucasians and African Americans.

Nonsteroidal anti-inflammatory drug use and Alzheimer's disease risk: the MIRAGE Study.

BACKGROUND: Nonsteroidal anti-inflammatory drugs (NSAID) use may protect against Alzheimer's disease (AD) risk. We sought examine the association between NSAID use and risk of AD, and potential effect modification by APOE-epsilon4 carrier status and ethnicity. METHODS: The MIRAGE Study is a multi-center family study of genetic and environmental risk factors for AD. Subjects comprised 691 AD patients (probands) and 973 family members enrolled at 15 research centers between 1996 and 2002. The primary independent and dependent variables were prior NSAID use and AD case status, respectively. We stratified the dataset in order to evaluate whether the association between NSAID use and AD was similar in APOE-epsilon4 carriers and non-carriers. Ethnicity was similarly examined as an effect modifier. RESULTS: NSAID use was less frequent in cases compared to controls in the overall sample (adjusted OR = 0.64; 95% CI = 0.38-1.05). The benefit of NSAID use appeared more pronounced among APOE-epsilon4 carriers (adjusted OR = 0.49; 95% CI = 0.24-0.98) compared to non-carriers, although this association was not statistically significant. The pattern of association was similar in Caucasian and African Americans. CONCLUSIONS: NSAID use is inversely associated with AD and may be modified by APOE genotype. Prospective studies and clinical trials of sufficient power to detect effect modification by APOE-epsilon4 carrier status are needed.

Differential modulation of endotoxin responsiveness by human caspase-12 polymorphisms.

Caspases mediate essential key proteolytic events in inflammatory cascades and the apoptotic cell death pathway. Human caspases functionally segregate into two distinct subfamilies: those involved in cytokine maturation (caspase-1, -4 and -5) and those involved in cellular apoptosis (caspase-2, -3, -6, -7, -8, -9 and -10). Although caspase-12 is phylogenetically related to the cytokine maturation caspases, in mice it has been proposed as a mediator of apoptosis induced by endoplasmic reticulum stress including amyloid-beta cytotoxicity, suggesting that it might contribute to the pathogenesis of Alzheimer's disease. Here we show that a single nucleotide polymorphism in caspase-12 in humans results in the synthesis of either a truncated protein (Csp12-S) or a full-length caspase proenzyme (Csp12-L). The read-through single nucleotide polymorphism encoding Csp12-L is confined to populations of African descent and confers hypo-responsiveness to lipopolysaccharide-stimulated cytokine production in ex vivo whole blood, but has no significant effect on apoptotic sensitivity. In a preliminary study, we find that the frequency of the Csp12-L allele is increased in African American individuals with severe sepsis. Thus, Csp12-L attenuates the inflammatory and innate immune response to endotoxins and in doing so may constitute a risk factor for developing sepsis.