Spike and Slab Gene Selection for Multigroup Microarray Data

Ishwaran, Hemant; Rao, J. Sunil

doi:10.1198/016214505000000051

Cited by 83 publications

(87 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Examples of diabetes studies with modest sample sizes include Lecker et al (35), who had a sample size of two normal and two diabetic animals; Suh et al (50), who had four normal and four diabetic animals at each age; Baelde et al (2), who had a sample size of two normal and two diabetic animals; and Fan et al (13), who had a sample size of two normal and three diabetic animals. Second, statistical methods designed for these types of studies, and in particular BAM, which was used in the present study, borrow strength across the data (27,28). That is, BAM takes advantage of the fact that there are thousands of genes whose changes are tested at the same time.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Oxidoreductase, morphogenesis, extracellular matrix, and calcium ion-binding gene expression in streptozotocin-induced diabetic rat heart

Lunteren

Moyer

2007

American Journal of Physiology-Endocrinology and Metabolism

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

“…bamarray.com) (27,28). BAM uses a special type of inferential regularization that borrows strength across the whole data set, thereby improving statistical power while at the same time reducing the false detection rate.…”

Section: Methodsmentioning

confidence: 99%

Oxidoreductase, morphogenesis, extracellular matrix, and calcium ion-binding gene expression in streptozotocin-induced diabetic rat heart

Lunteren

Moyer

2007

American Journal of Physiology-Endocrinology and Metabolism

View full text Add to dashboard Cite

show abstract

“…The hyperparameters for the scaled inverse chisquare priors in this study are chosen in a subjective way as done in Ishwaran and Rao (2005). According to Ishwaran and Rao (2003), these values do not need to be tuned for each data set and can be fixed.…”

Section: Applicationsmentioning

confidence: 99%

Mapping Quantitative Trait Loci for Expression Abundance

Jia

2007

Genetics

View full text Add to dashboard Cite

Mendelian loci that control the expression levels of transcripts are called expression quantitative trait loci (eQTL). When mapping eQTL, we often deal with thousands of expression traits simultaneously, which complicates the statistical model and data analysis. Two simple approaches may be taken in eQTL analysis: (1) individual transcript analysis in which a single expression trait is mapped at a time and the entire eQTL mapping involves separate analysis of thousands of traits and (2) individual marker analysis where differentially expressed transcripts are detected on the basis of their association with the segregation pattern of an individual marker and the entire analysis requires scanning markers of the entire genome. Neither approach is optimal because data are not analyzed jointly. We develop a Bayesian clustering method that analyzes all expressed transcripts and markers jointly in a single model. A transcript may be simultaneously associated with multiple markers. Additionally, a marker may simultaneously alter the expression of multiple transcripts. This is a model-based method that combines a Gaussian mixture of expression data with segregation of multiple linked marker loci. Parameter estimation for each variable is obtained via the posterior mean drawn from a Markov chain Monte Carlo sample. The method allows a regular quantitative trait to be included as an expression trait and subject to the same clustering assignment. If an expression trait links to a locus where a quantitative trait also links, the expressed transcript is considered to be associated with the quantitative trait. The method is applied to a microarray experiment with 60 F 2 mice measured for 25 different obesity-related quantitative traits. In the experiment, $40,000 transcripts and 145 codominant markers are investigated for their associations. A program written in SAS/IML is available from the authors on request.

show abstract

“…One advantage with the BAMarray approach is therefore increased power. (23) For identifying differentially expressed genes, a Bayesian test statistic is estimated for each gene. Contrary to most other software packages, which require a user-specified cutoff value for the test statistic, BAMarray determines the cutoff value automatically from the data to obtain an optimal balance between total false detections and total false nondetections.…”

Section: Biochemical Parametersmentioning

confidence: 99%

Molecular disease map of bone characterizing the postmenopausal osteoporosis phenotype

Jemtland

Holden

Reppe

et al. 2011

Journal of Bone and Mineral Research

View full text Add to dashboard Cite

Genome-wide gene expressions in bone biopsies from patients with postmenopausal osteoporosis and healthy controls were profiled, to identify osteoporosis candidate genes. All osteoporotic patients (n ¼ 27) in an unbiased cohort of Norwegian women presented with bone mineral density (BMD) T-scores of less than À2.5 SD and one or more confirmed low-energy fracture(s). A validation group (n ¼ 18) had clinical and laboratory parameters intermediate to the control (n ¼ 39) and osteoporosis groups. RNA from iliac crest bone biopsies were analyzed by Affymetrix microarrays and real-time reverse-transcriptase polymerase chain reaction (RT-PCR). Differentially expressed genes in osteoporosis versus control groups were identified using the Bayesian ANOVA for microarrays (BAMarray) method, whereas the R-package Limma (Linear Models for Microarray Data) was used to determine whether these transcripts were explained by disease, age, body mass index (BMI), or combinations thereof. Laboratory tests showed normal ranges for the cohort. A total of 609 transcripts were differentially expressed in osteoporotic patients relative to controls; 256 transcripts were confirmed for disease when controlling for age or BMI. Most of the osteoporosis susceptibility genes (80%) also were confirmed to be regulated in the same direction in the validation group. Furthermore, 217 of 256 transcripts were correlated with BMD (adjusted for age and BMI) at various skeletal sites (jrj > 0.2, p < .05). Among the most distinctly expressed genes were Wnt antagonists DKK1 and SOST, the transcription factor SOX4, and the bone matrix proteins MMP13 and MEPE, all reduced in osteoporosis versus control groups. Our results identify potential osteoporosis susceptibility candidate genes adjusted for confounding factors (ie, age and BMI) with or without a significant correlation with BMD. ß

show abstract

Spike and Slab Gene Selection for Multigroup Microarray Data

Cited by 83 publications

References 14 publications

Oxidoreductase, morphogenesis, extracellular matrix, and calcium ion-binding gene expression in streptozotocin-induced diabetic rat heart

Oxidoreductase, morphogenesis, extracellular matrix, and calcium ion-binding gene expression in streptozotocin-induced diabetic rat heart

Mapping Quantitative Trait Loci for Expression Abundance

Molecular disease map of bone characterizing the postmenopausal osteoporosis phenotype

Contact Info

Product

Resources

About