Methylation-eQTL analysis in cancer research.

MOTIVATION: DNA methylation is a key epigenetic factor regulating gene expression. While promoter methylation has been well studied, recent publications have revealed that functionally important methylation also occurs in intergenic and distal regions, and varies across genes and tissue types. Given the growing importance of inter-platform integrative genomic analyses, there is an urgent need to develop methods to discover and characterize gene-level relationships between methylation and expression. RESULTS: We introduce a novel sequential penalized regression approach to identify methylation-expression quantitative trait loci (methyl-eQTLs), a term that we have coined to represent, for each gene and tissue type, a sparse set of CpG loci best explaining gene expression and accompanying weights indicating direction and strength of association. Using TCGA and MD Anderson colorectal cohorts to build and validate our models, we demonstrate our strategy better explains expression variability than current commonly used gene-level methylation summaries. The methyl-eQTLs identified by our approach can be used to construct gene-level methylation summaries that are maximally correlated with gene expression for use in integrative models, and produce a tissue-specific summary of which genes appear to be strongly regulated by methylation. Our results introduce an important resource to the biomedical community for integrative genomics analyses involving DNA methylation. AVAILABILITY AND IMPLEMENTATION: We produce an R Shiny app (https://rstudio-prd-c1.pmacs.upenn.edu/methyl-eQTL/) that interactively presents methyl-eQTL results for colorectal, breast and pancreatic cancer. The source R code for this work is provided in the Supplementary Material. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Proteomic Profiling of Signaling Networks Modulated by G-CSF/Plerixafor/Busulfan-Fludarabine Conditioning in Acute Myeloid Leukemia Patients in Remission or with Active Disease prior to Allogeneic Stem Cell Transplantation.

To characterize intracellular signaling in peripheral blood (PB) cells of acute myeloid leukemia (AML) patients undergoing pretransplant conditioning with CXCR4 inhibitor plerixafor, granulocyte colony-stimulating factor (G-CSF), and busulfan plus fludarabine (Bu+Flu) chemotherapy, we profiled 153 proteins in 33 functional groups using reverse phase protein array. CXCR4 inhibition mobilized AML progenitors and clonal AML cells, and this was associated with molecular markers of cell cycle progression. G-CSF/plerixafor and G-CSF/plerixafor/Bu+Flu modulated distinct signaling networks in AML blasts of patients undergoing conditioning with active disease compared to nonleukemic PB cells of patients in remission. We identified AML-specific proteins that remained aberrantly expressed after chemotherapy, representing putative chemoresistance markers in AML.

Genetic Polymorphisms of CFH and ARMS2 Do Not Predict Response to Antioxidants and Zinc in Patients with Age-Related Macular Degeneration: Independent Statistical Evaluations of Data from the Age-Related Eye Disease Study.

PURPOSE: Considerable controversy has erupted in recent years regarding whether genotyping should be part of standard care for patients with age-related macular degeneration (AMD) who are being considered for treatment with antioxidants and zinc. We aimed to determine whether genotype predicts response to supplements in AMD. DESIGN: Three separate statistical teams reanalyzed data derived from the Age-Related Eye Disease Study (AREDS), receiving data prepared by the AREDS investigators and, separately, data from investigators reporting findings that support the use of genotyping. PARTICIPANTS: The population of interest was AREDS participants with AMD worse than category 1 and genotyping data available. Data from the 2 groups overlap imperfectly with respect to measurements made: the largest common set involved 879 participants for whom the same CFH and ARMS2 single nucleotide polymorphisms were measured by both groups. METHODS: Each team took a separate but complementary approach. One team focused on data concordance between conflicting studies. A second team focused on replicating the key claim of an interaction between genotype and treatment. The third team took a blank slate approach in attempting to find baseline predictors of treatment response. MAIN OUTCOME MEASURES: Progression to advanced AMD. RESULTS: We found errors in the data used to support the initial claim of genotype-treatment interaction. Although we found evidence that high-risk patients had more to gain from treatment, we were unable to replicate any genotype-treatment interactions after adjusting for multiple testing. We tested 1 genotype claim on an independent set of data, with negative results. Even if we assumed that interactions in fact did exist, we did not find evidence to support the claim that supplementation leads to a large increase in the risk of advanced AMD in some genotype subgroups. CONCLUSIONS: Patients who meet criteria for supplements to prevent AMD progression should be offered zinc and antioxidants without consideration of genotype.

High-throughput profiling of signaling networks identifies mechanism-based combination therapy to eliminate microenvironmental resistance in acute myeloid leukemia.

The bone marrow microenvironment is known to provide a survival advantage to residual acute myeloid leukemia cells, possibly contributing to disease recurrence. The mechanisms by which stroma in the microenvironment regulates leukemia survival remain largely unknown. Using reverse-phase protein array technology, we profiled 53 key protein molecules in 11 signaling pathways in 20 primary acute myeloid leukemia samples and two cell lines, aiming to understand stroma-mediated signaling modulation in response to the targeted agents temsirolimus (MTOR), ABT737 (BCL2/BCL-XL), and Nutlin-3a (MDM2), and to identify the effective combination therapy targeting acute myeloid leukemia in the context of the leukemia microenvironment. Stroma reprogrammed signaling networks and modified the sensitivity of acute myeloid leukemia samples to all three targeted inhibitors. Stroma activated AKT at Ser473 in the majority of samples treated with single-agent ABT737 or Nutlin-3a. This survival mechanism was partially abrogated by concomitant treatment with temsirolimus plus ABT737 or Nutlin-3a. Mapping the signaling networks revealed that combinations of two inhibitors increased the number of affected proteins in the targeted pathways and in multiple parallel signaling, translating into facilitated cell death. These results demonstrated that a mechanism-based selection of combined inhibitors can be used to guide clinical drug selection and tailor treatment regimens to eliminate microenvironment-mediated resistance in acute myeloid leukemia.

Integrated MicroRNA-mRNA Profiling Identifies Oncostatin M as a Marker of Mesenchymal-Like ER-Negative/HER2-Negative Breast Cancer.

MicroRNAs (miRNAs) simultaneously modulate different oncogenic networks, establishing a dynamic system of gene expression and pathway regulation. In this study, we analyzed global miRNA and messenger RNA (mRNA) expression profiles of 17 cell lines representing different molecular breast cancer subtypes. Spearman's rank correlation test was used to evaluate the correlation between miRNA and mRNA expression. Hierarchical clustering and pathway analysis were also performed. Publicly available gene expression profiles (n = 699) and tumor tissues (n = 80) were analyzed to assess the relevance of key miRNA-regulated pathways in human breast cancer. We identified 39 significantly deregulated miRNAs, and the integration between miRNA and mRNA data revealed the importance of immune-related pathways, particularly the Oncostatin M (OSM) signaling, associated with mesenchymal-like breast cancer cells. OSM levels correlated with genes involved in the inflammatory response, epithelial-to-mesenchymal transition (EMT), and epidermal growth factor (EGF) signaling in human estrogen receptor (ER)-negative/human epidermal growth factor receptor 2 (HER2)-negative breast cancer. Our results suggest that the deregulation of specific miRNAs may cooperatively impair immune and EMT pathways. The identification of the OSM inflammatory pathway as an important mediator of EMT in triple-negative breast cancer (TNBC) may provide a novel potential opportunity to improve therapeutic strategies.

A genome-scale screen reveals context-dependent ovarian cancer sensitivity to miRNA overexpression.

Large-scale molecular annotation of epithelial ovarian cancer (EOC) indicates remarkable heterogeneity in the etiology of that disease. This diversity presents a significant obstacle against intervention target discovery. However, inactivation of miRNA biogenesis is commonly associated with advanced disease. Thus, restoration of miRNA activity may represent a common vulnerability among diverse EOC oncogenotypes. To test this, we employed genome-scale, gain-of-function, miRNA mimic toxicity screens in a large, diverse spectrum of EOC cell lines. We found that all cell lines responded to at least some miRNA mimics, but that the nature of the miRNA mimics provoking a response was highly selective within the panel. These selective toxicity profiles were leveraged to define modes of action and molecular response indicators for miRNA mimics with tumor-suppressive characteristics in vivo. A mechanistic principle emerging from this analysis was sensitivity of EOC to miRNA-mediated release of cell fate specification programs, loss of which may be a prerequisite for development of this disease.

Intratumoral morphologic and molecular heterogeneity of rhabdoid renal cell carcinoma: challenges for personalized therapy.

Rhabdoid histology in clear-cell renal cell carcinoma is associated with a poor prognosis. The prognosis of patients with clear-cell renal cell carcinoma may also be influenced by molecular alterations. The aim of this study was to evaluate the association between histologic features and salient molecular changes in rhabdoid clear-cell renal cell carcinoma. We macrodissected the rhabdoid and clear-cell epithelioid components from 12 cases of rhabdoid clear-cell renal cell carcinoma. We assessed cancer-related mutations from eight cases using a clinical next-generation exome-sequencing platform. The transcriptome of rhabdoid clear-cell renal cell carcinoma (n=8) and non-rhabdoid clear-cell renal cell carcinoma (n=37) was assessed by RNA-seq and gene expression microarray. VHL (63%) showed identical mutations in all regions from the same tumor. BAP1 (38%) and PBRM1 (13%) mutations were identified in the rhabdoid but not in the epithelioid component and were mutually exclusive in 3/3 cases and 1 case, respectively. SETD2 (63%) mutations were discordant between different histologic regions in 2/5 cases, with mutations called only in the epithelioid and rhabdoid components, respectively. The transcriptome of rhabdoid clear-cell renal cell carcinoma was distinct from advanced-stage and high-grade clear-cell renal cell carcinoma. The diverse histologic components of rhabdoid clear-cell renal cell carcinoma, however, showed a similar transcriptomic program, including a similar prognostic gene expression signature. Rhabdoid clear-cell renal cell carcinoma is transcriptomically distinct and shows a high rate of SETD2 and BAP1 mutations and a low rate of PBRM1 mutations. Driver mutations in clear-cell renal cell carcinoma are often discordant across different morphologic regions, whereas the gene expression program is relatively stable. Molecular profiling of clear-cell renal cell carcinoma may improve by assessing for gene expression and sampling tumor foci from different histologic regions.

Reverse-phase protein array for prediction of patients at low risk of developing bone metastasis from breast cancer.

BACKGROUND: A biomarker that predicts bone metastasis based on a protein laboratory assay has not been demonstrated. Reverse-phase protein array (RPPA) enables quantification of total and phosphorylated proteins, providing information about their functional status. The aim of this study was to identify bone-metastasis-related markers in patients with primary breast cancer using RPPA analysis. PATIENTS AND METHODS: Tumor samples were obtained from 169 patients with primary invasive breast carcinoma who underwent surgery. The patients were categorized by whether they developed breast cancer bone metastasis (BCBM) during follow-up. Clinical characteristics and protein expression by RPPA were compared and verified by leave-one-out cross-validation. RESULTS: Lymph node status (p = .023) and expression level of 22 proteins by RPPA were significantly correlated with BCBM in logistic regression analysis. These variables were used to build a logistic regression model. After filtering the variables through a stepwise algorithm, the final model, consisting of 8 proteins and lymph node status, had sensitivity of 30.0%, specificity of 90.5%, positive predictive value of 30.0%, and negative predictive value of 90.5% in the cross-validation. Most of the identified proteins were associated with cell cycle or signal transduction (CDK2, CDKN1A, Rb1, Src, phosphorylated-ribosomal S6 kinase, HER2, BCL11A, and MYH11). CONCLUSION: Our validated model, in which the primary tumor is tested with RPPA, can predict patients who are at low risk of developing BCBM and thus who likely would not benefit from receiving a bisphosphonate in the adjuvant setting. Clinical trials excluding these patients have the potential to clarify the benefit of bisphosphonates in the adjuvant setting.

Targeting of mTORC1/2 by the mTOR kinase inhibitor PP242 induces apoptosis in AML cells under conditions mimicking the bone marrow microenvironment.

The interactions between the bone marrow (BM) microenvironment and acute myeloid leukemia (AML) is known to promote survival of AML cells. In this study, we used reverse phase-protein array (RPPA) technology to measure changes in multiple proteins induced by stroma in leukemic cells. We then investigated the potential of an mTOR kinase inhibitor, PP242, to disrupt leukemia/stroma interactions, and examined the effects of PP242 in vivo using a mouse model. Using RPPA, we confirmed that multiple survival signaling pathways, including the phosphatidylinositol 3-kinase (PI3K)/protein kinase B (AKT)/mammalian target of rapamycin (mTOR), were up-regulated in primary AML cells cocultured with stroma. PP242 effectively induced apoptosis in primary samples cultured with or without stroma. Mechanistically, PP242 attenuated the activities of mTORC1 and mTORC2, sequentially inhibited phosphorylated AKT, S6K, and 4EBP1, and concurrently suppressed chemokine receptor CXCR4 expression in primary leukemic cells and in stromal cells cultured alone or cocultured with leukemic cells. In the in vivo leukemia mouse model, PP242 inhibited mTOR signaling in leukemic cells and demonstrated a greater antileukemia effect than rapamycin. Our findings indicate that disrupting mTOR/AKT signaling with a selective mTOR kinase inhibitor can effectively target leukemic cells within the BM microenvironment.

Comparison of molecular subtype distribution in triple-negative inflammatory and non-inflammatory breast cancers.

INTRODUCTION: Because of its high rate of metastasis, inflammatory breast cancer (IBC) has a poor prognosis compared with non-inflammatory types of breast cancer (non-IBC). In a recent study, Lehmann and colleagues identified seven subtypes of triple-negative breast cancer (TNBC). We hypothesized that the distribution of TNBC subtypes differs between TN-IBC and TN-non-IBC. We determined the subtypes and compared clinical outcomes by subtype in TN-IBC and TN-non-IBC patients. METHODS: We determined TNBC subtypes in a TNBC cohort from the World IBC Consortium for which IBC status was known (39 cases of TN-IBC; 49 cases of TN-non-IBC). We then determined the associations between TNBC subtypes and IBC status and compared clinical outcomes between TNBC subtypes. RESULTS: We found the seven subtypes exist in both TN-IBC and TN-non-IBC. We found no association between TNBC subtype and IBC status (P = 0.47). TNBC subtype did not predict recurrence-free survival. IBC status was not a significant predictor of recurrence-free or overall survival in the TNBC cohort. CONCLUSIONS: Our data show that, like TN-non-IBC, TN-IBC is a heterogeneous disease. Although clinical characteristics differ significantly between IBC and non-IBC, no unique IBC-specific TNBC subtypes were identified by mRNA gene-expression profiles of the tumor. Studies are needed to identify the subtle molecular or microenvironmental differences that contribute to the differing clinical behaviors between TN-IBC and TN-non-IBC.

Epigenetic analysis of the Notch superfamily in high-grade serous ovarian cancer.

OBJECTIVES: Gene methylation and other epigenetic modifications of gene regulation have been implicated in the growth of ovarian cancer, but the clinical significance of such modifications in the Notch pathway in high-grade serous ovarian cancer (HGS-OvCa) is not well understood. We used The Cancer Genome Atlas (TCGA) data to study the clinical relevance of epigenetic modifications of Notch superfamily genes. METHODS: We analyzed the interaction of DNA methylation and miRNAs with gene expression data for Notch superfamily members with the Spearman rank correlation test and explored potential relationships with overall survival (OS) with the log-rank test. We downloaded clinical data, level 3 gene expression data, and level 3 DNA methylation data for 480 patients with stage II-IV HGS-OvCa from the TCGA data portal. Patients were randomly divided into training and validation cohorts for survival analyses. In each set, patients were grouped into percentiles according to methylation and microRNA (miRNA) or messenger RNA (mRNA) levels. We used several algorithms to predict miRNA-mRNA interaction. RESULTS: There were significant inverse relationships between methylation status and mRNA expression for PPARG, CCND1, and RUNX1. For each of these genes, patients with a lower methylation level and higher expression level had significantly poorer OS than did patients with a higher methylation level and lower expression level. We also found a significant inverse relationship between miRNAs and mRNA expression for CCND1, PPARG, and RUNX1. By further analyzing the effect of miRNAs on gene expression and OS, we found that patients with higher levels of CCND1, PPARG, and RUNX1 expression and lower expression levels of their respective miRNAs (502-5p, 128, and 215/625) had significantly poorer OS. CONCLUSIONS: Epigenetic alterations of multiple Notch target genes and pathway interacting genes (PPARG, CCND1, and RUNX1) may relate to activation of this pathway and poor survival of patients with HGS-OvCa.

Protein and phosphoprotein levels in glioma and adenocarcinoma cell lines grown in normoxia and hypoxia in monolayer and three-dimensional cultures.

BACKGROUND: Three dimensional (3D) growths of cancer cells in vitro are more reflective of in situ cancer cell growth than growth in monolayer (2D). The present study is designed to determine changes in protein and phosphoprotein that reflect adaptation of tumor cells to 3D as compared to 2D. Since relative hypoxia is a common feature of most solid tumors, the present study also aims to look at the impact of transition from normoxia to hypoxia in these two growth conditions. RESULTS: Using reverse-phase protein arrays, we compared levels of 121 different phosphorylated and non-phosphorylated proteins in 5 glioma and 6 adenocarcinoma lines under conditions of 3D and monolayer culture in normoxia and hypoxia. A three-way analysis of variance showed levels of 82 antibodies differed between media (2D vs. 3D) and 49 differed between treatments (hypoxia vs. normoxia). Comparing 2D to 3D growth, 7 proteins were commonly (i.e., > 50% of tumors) elevated in 3D: FAK, AKT, Src, GSK3αß, TSC2, p38, and NFκßp65. Conversely, 7 other proteins are commonly decreased: ATRIP, ATR, ß-catenin, BCL-X, cyclin B1, Egr-1, and HIF-1α. Comparing normoxia to hypoxia, only NCKIPSD was commonly elevated in hypoxia; 6 proteins were decreased: cyclin B1, 4EBP1(Ser65), c-Myc, SMAD3(Ser423), S6(Ser235), and S6(Ser240). Hypoxia affected glioma cell lines differently from adenocarcinoma cell lines: 8 proteins were increased in gliomas (BAX, caspase 7, HIF-1α, c-JUN, MEK1, PARP 1 cleaved, Src, and VEGFR2) and none in adenocarcinomas. CONCLUSIONS: We identified subsets of proteins with clearly concordant/discordant behavior between gliomas and adenocarcinomas. In general, monolayer to 3D culture differences are clearer than normoxia to hypoxia differences, with anti-apoptotic, cytoskeletal rearrangement and cell survival pathways emphasized in the former and mTOR pathway, transcription, cell-cycle arrest modulation, and increased cell motility in the latter.

Molecular Analysis of Short- versus Long-Term Survivors of High-Grade Serous Ovarian Carcinoma.

Despite having similar histologic features, patients with high-grade serous ovarian carcinoma (HGSC) often experience highly variable outcomes. The underlying determinants for long-term survival (LTS, ≥10 years) versus short-term survival (STS, <3 years) are largely unknown. The present study sought to identify molecular predictors of LTS for women with HGSC. A cohort of 24 frozen HGSC samples was collected (12 LTS and 12 STS) and analyzed at DNA, RNA, and protein levels. OVCAR5 and OVCAR8 cell lines were used for in vitro validation studies. For in vivo studies, we injected OVCAR8 cells into the peritoneal cavity of female athymic nude mice. From RNAseq analysis, 11 genes were found to be differentially expressed between the STS and LTS groups (fold change > 2; false discovery rate < 0.01). In the subsequent validation cohort, transmembrane protein 62 (TMEM62) was found to be related to LTS. CIBERSORT analysis showed that T cells (follicular helper) were found at higher levels in tumors from LTS than STS groups. In vitro data using OVCAR5 and OVCAR8 cells showed decreased proliferation with TMEM62 overexpression and positive correlation with a longevity-regulating pathway (KEGG HSA04213) at the RNA level. In vivo analysis using the OVCAR8-TMEM62-TetON model showed decreased tumor burden in mice with high- vs. low-expressing TMEM62 tumors. Our results demonstrate that restoring TMEM62 may be a novel approach for treatment of HGSC. These findings may have implications for biomarker and intervention strategies to help improve patient outcomes

Steps toward mapping the human vasculature by phage display.

The molecular diversity of receptors in human blood vessels remains largely unexplored. We developed a selection method in which peptides that home to specific vascular beds are identified after administration of a peptide library. Here we report the first in vivo screening of a peptide library in a patient. We surveyed 47,160 motifs that localized to different organs. This large-scale screening indicates that the tissue distribution of circulating peptides is nonrandom. High-throughput analysis of the motifs revealed similarities to ligands for differentially expressed cell-surface proteins, and a candidate ligand-receptor pair was validated. These data represent a step toward the construction of a molecular map of human vasculature and may have broad implications for the development of targeted therapies.

Gene Body Methylation of the Lymphocyte-Specific Gene CARD11 Results in Its Overexpression and Regulates Cancer mTOR Signaling.

Investigations into the function of nonpromoter DNA methylation have yielded new insights into epigenetic regulation of gene expression. Previous studies have highlighted the importance of distinguishing between DNA methylation in discrete functional regions; however, integrated nonpromoter DNA methylation and gene expression analyses across a wide number of tumor types and corresponding normal tissues have not been performed. Through integrated analysis of gene expression and DNA methylation profiles, we examined 32 tumor types and identified 57 tumor suppressors and oncogenes out of 260 genes exhibiting a correlation of > 0.5 between gene body methylation and gene expression in at least one tumor type. The lymphocyte-specific gene CARD11 exhibits robust association between gene body methylation and expression across 19 of 32 tumor types examined. It is significantly overexpressed in kidney renal cell carcinoma (KIRC) and lung adenocarcinoma (LUAD) tumor tissues in comparison with respective control samples; and is significantly associated with lower overall survival in KIRC. Contrary to its canonical function in lymphocyte NFκB activation, CARD11 activates the mTOR pathway in KIRC and LUAD, resulting in suppressed autophagy. Furthermore, demethylation of a CpG island within the gene body of CARD11 decreases gene expression. Collectively, our study highlights how DNA methylation outside the promoter region can impact tumor progression. IMPLICATIONS: Our study describes a novel regulatory role of gene body DNA methylation-dependent CARD11 expression on mTOR signaling and its impact on tumor progression.

Different changes in protein and phosphoprotein levels result from serum starvation of high-grade glioma and adenocarcinoma cell lines.

Tumor cells undergoing serum starvation in vitro partially mimic metabolically stressed cells trying to adjust to a changed environment in vivo by inducing signal transduction and gene expression so that the tumor continues to grow. Our hypothesis is that the changes in protein and phosphoprotein levels after serum starvation may reflect the adapted phenotype of the tumor, which could be targeted for therapy. We used reverse-phase protein microarrays to interrogate five high-grade glioma cell lines and seven adenocarcinoma cell lines for differences in the level of 81 proteins and 25 phosphoproteins. All cell lines were studied in the well-fed condition of growth with 10% FBS and the starved condition of 0.5% FBS. Protein expression levels were normalized to beta-actin and trichotomized as increased (+1, upper 75th quartile), decreased (-1, lowest 25th quartile), or unchanged (0, others) to focus on the patterns of the biggest (and hopefully most robust) changes in protein and phosphoprotein levels. We examined these trichotomized values to better understand Starved-Fed differences among the cell lines and thereby gain better/clearer insight into the effects of serum starvation on potential cellular responses. In general, the expression of proteins and phosphoproteins 24 h after FBS starvation increased more often in glioma lines than in adenocarcinoma lines, which appeared to have fewer increased protein scores and more decreased scores. Many of the proteins increased in gliomas were downstream targets of the PTEN-PI-3 kinase-AKT, EGFR-MAPK-Stat, and transcription activator-polyamine signaling pathways. In adenocarcinomas, the expression of proteins and phosphoproteins generally increased in apoptosis pathways, while there were minor fluctuations in the other pathways above. Contrawise, gliomas become resistant to apoptosis after 24 h of serum starvation and upregulate transcription activators and polyamines more so than adenocarciomas.

Clinical evaluation of chemotherapy response predictors developed from breast cancer cell lines.

The goal of this study was to develop pharmacogenomic predictors in response to standard chemotherapy drugs in breast cancer cell lines and test their predictive value in patients who received treatment with the same drugs. Nineteen human breast cancer cell lines were tested for sensitivity to paclitaxel (T), 5-fluorouracil (F), doxorubicin (A) and cyclophosphamide (C) in vitro. Baseline gene expression data were obtained for each cell line with Affymetrix U133A gene chips, and multigene predictors of sensitivity were derived for each drug separately. These predictors were applied individually and in combination to human gene expression data generated with the same Affymetrix platform from fine needle aspiration specimens of 133 stage I-III breast cancers. Tumor samples were obtained at baseline, and each patient received 6 months of preoperative TFAC chemotherapy followed by surgery. Cell line-derived prediction results were correlated with the observed pathologic response to chemotherapy. Statistically robust differentially expressed genes between sensitive and resistant cells could only be found for paclitaxel. False discovery rates associated with the informative genes were high for all other drugs. For each drug, the top 100 differentially expressed genes were combined into a drug-specific response predictor. When these cell line-based predictors were applied to patient data, there was no significant correlation between observed response and predicted response either for individual drug predictors or combined predictions. Cell line-derived predictors of response to four commonly used chemotherapy drugs did not predict response accurately in patients.

Variable slope normalization of reverse phase protein arrays.

MOTIVATION: Reverse phase protein arrays (RPPA) measure the relative expression levels of a protein in many samples simultaneously. A set of identically spotted arrays can be used to measure the levels of more than one protein. Protein expression within each sample on an array is estimated by borrowing strength across all the samples, but using only within array information. When comparing across slides, it is essential to account for sample loading, the total amount of protein printed per sample. Currently, total protein is estimated using either a housekeeping protein or the sample median across all slides. When the variability in sample loading is large, these methods are suboptimal because they do not account for the fact that the protein expression for each slide is estimated separately. RESULTS: We propose a new normalization method for RPPA data, called variable slope (VS) normalization, that takes into account that quantification of RPPA slides is performed separately. This method is better able to remove loading bias and recover true correlation structures between proteins. AVAILABILITY: Code to implement the method in the statistical package R and anonymized data are available at (http://bioinformatics.mdanderson.org/supplements.html).

Differences in aberrant expression and splicing of sarcomeric proteins in the myotonic dystrophies DM1 and DM2.

Aberrant transcription and mRNA processing of multiple genes due to RNA-mediated toxic gain-of-function has been suggested to cause the complex phenotype in myotonic dystrophies type 1 and 2 (DM1 and DM2). However, the molecular basis of muscle weakness and wasting and the different pattern of muscle involvement in DM1 and DM2 are not well understood. We have analyzed the mRNA expression of genes encoding muscle-specific proteins and transcription factors by microarray profiling and studied selected genes for abnormal splicing. A subset of the abnormally regulated genes was further analyzed at the protein level. TNNT3 and LDB3 showed abnormal splicing with significant differences in proportions between DM2 and DM1. The differential abnormal splicing patterns for TNNT3 and LDB3 appeared more pronounced in DM2 relative to DM1 and are among the first molecular differences reported between the two diseases. In addition to these specific differences, the majority of the analyzed genes showed an overall increased expression at the mRNA level. In particular, there was a more global abnormality of all different myosin isoforms in both DM1 and DM2 with increased transcript levels and a differential pattern of protein expression. Atrophic fibers in DM2 patients expressed only the fast myosin isoform, while in DM1 patients they co-expressed fast and slow isoforms. However, there was no increase of total myosin protein levels, suggesting that aberrant protein translation and/or turnover may also be involved.