Citations

This article generalizes a multivariate skew-normal distribution and describes its many interesting properties. The univariate version of the new distribution is compared with two other currently used distributions. The use of the new distribution is illustrated with a real data example suitable for regression modeling. The new model provides a better model than its two rivals as evaluated by some suitable Bayesian model selection criteria. AMS (2000) Subject Classification : 62F15, 62J05.

Abstract Background: We investigated the association of germline genetic variation in hormone biosynthesis and metabolism genes with prostate cancer specific mortality in a cohort of advanced prostate cancer patients. Methods: We genotyped 852 single nucleotide polymorphisms (SNPs) from 97 genes in a cohort of 267 advanced prostate cancer patients at the time of progression to castration recurrence (CRPC) during on-going androgen ablation. Tagging SNPs with minor allele frequency (MAF) of >5% and r2 ≥0.8 were selected from HapMap, NIEHS and Seattle SNP databases. Medical records were queried for cause of death. The primary endpoint of time to prostate cancer specific mortality (PCSM), was pre-defined as time from development of CRPC to death from prostate cancer progression. Principle components analysis was used for gene-levels tests, and to account for multiple testing, we calculated the false discovery rate (FDR). For SNP level results, Hazard ratios (HR) and 95% confidence intervals (CI) were estimated using cox regression. Results: The median age of the cohort was 72 years at CRPC. 43% had a Gleason score (GS)=8-10, 33% a GS=7, and 24% a GS<7. After a median follow up of 1.8 years (IQ range: 0.8-3.3 years), 139 patients died, of which 107 were due to prostate cancer progression. In unadjusted gene level analyses, UGT1A7 (p=0.0059; FDR=0.19), UGT1A10 (p=0.0017; FDR=0.17) and UGT1A3 (p=0.0037; FDR=0.18) were associated with PCSM. After adjusting for age and GS, SNPs strongly associated with PCSM are listed in the Table. Conclusions: Variation in UGT genes involved in hormone metabolism yield prognostic information in CRPC. Further validation is needed to develop these as prognostic biomarkers. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 4174. doi:10.1158/1538-7445.AM2011-4174

Abstract Motivation: Overexpression of HER2 (the product of the ERBB2 gene) occurs in about 15% of all breast tumors. We have undertaken to use next generation transcriptome sequencing technology to identify genomic features that are unique to HER2+ tumors. Interactome mapping was then used to integrate the genes associated with these features into a transcriptome landscape model, with a view towards identifying nodes of interaction that might be targeted in HER2+ tumors. Methods: We performed 50nt paired-end RNA-sequencing for 24 breast tumors: 8 each HER2+, ER+, triple negatives (TN). In addition to breast adenocarcinomas, we also sequenced 8 early passage non-transformed HMEC cell lines as normal controls. Reads from RNA sequencing were aligned to the genome and exon junctions using TopHat software. Gene counts were summarized and annotations were performed using our in-house programs. Tukey's test was used to obtain genes or transcripts that are specific to HER2 tumor group compared to other tumors/normal. A combination of bioinformatics softwares and algorithms were used to identify SNVs. Results: Only 13527 genes with median gene count greater than 16 reads in at least one of the 4 groups were considered for differential gene expression and splicing analysis. Some 696 genes were differentially expressed in HER2+ tumors compared to ER+, TN tumors and HMECs. We identified 272 alternately spliced transcripts for which the HER2+ tumors exhibited a mean transcript expression ratio significantly different from the means of other tumor groups. We also identified 4735 expressed single nucleotide variants that are uniquely associated with HER2+ tumors compared to other tumors/normal groups. Among these 3579/4735 sequence polymorphisms were not present in the 1000 genome germline sequence database or in the dbSNP132 database of naturally occurring germline polymorphisms. Integration of all the genes obtained from genomic feature analyses (differential expression, alternative splicing, single nucleotide variance) has been carried out to indentify biological processes that are specific to the HER2+ tumor subtype. Key nodes and pathways that are specific to HER2+ tumors will be evaluated for association with clinical outcome in a large series of patients who have received HER2-targeted therapy. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 4926. doi:1538-7445.AM2012-4926