Yingtao Bi

Imaging patterns predict patient survival and molecular subtype in glioblastoma via machine learning techniques.

BACKGROUND MRI characteristics of brain gliomas have been used to predict clinical outcome and molecular tumor characteristics. However, previously reported imaging biomarkers have not been sufficiently accurate or reproducible to enter routine clinical practice and often rely on relatively simple MRI measures. The current study leverages advanced image analysis and machine learning algorithms to identify complex and reproducible imaging patterns predictive of overall survival and molecular subtype in glioblastoma (GB). METHODS One hundred five patients with GB were first used to extract approximately 60 diverse features from preoperative multiparametric MRIs. These imaging features were used by a machine learning algorithm to derive imaging predictors of patient survival and molecular subtype. Cross-validation ensured generalizability of these predictors to new patients. Subsequently, the predictors were evaluated in a prospective cohort of 29 new patients. RESULTS Survival curves yielded a hazard ratio of 10.64 for predicted long versus short survivors. The overall, 3-way (long/medium/short survival) accuracy in the prospective cohort approached 80%. Classification of patients into the 4 molecular subtypes of GB achieved 76% accuracy. CONCLUSIONS By employing machine learning techniques, we were able to demonstrate that imaging patterns are highly predictive of patient survival. Additionally, we found that GB subtypes have distinctive imaging phenotypes. These results reveal that when imaging markers related to infiltration, cell density, microvascularity, and blood-brain barrier compromise are integrated via advanced pattern analysis methods, they form very accurate predictive biomarkers. These predictive markers used solely preoperative images, hence they can significantly augment diagnosis and treatment of GB patients.

HINCUTs in cancer: hypoxia-induced noncoding ultraconserved transcripts

Recent data have linked hypoxia, a classic feature of the tumor microenvironment, to the function of specific microRNAs (miRNAs); however, whether hypoxia affects other types of noncoding transcripts is currently unknown. Starting from a genome-wide expression profiling, we demonstrate for the first time a functional link between oxygen deprivation and the modulation of long noncoding transcripts from ultraconserved regions, termed transcribed-ultraconserved regions (T-UCRs). Interestingly, several hypoxia-upregulated T-UCRs, henceforth named ‘hypoxia-induced noncoding ultraconserved transcripts’ (HINCUTs), are also overexpressed in clinical samples from colon cancer patients. We show that these T-UCRs are predominantly nuclear and that the hypoxia-inducible factor (HIF) is at least partly responsible for the induction of several members of this group. One specific HINCUT, uc.475 (or HINCUT-1) is part of a retained intron of the host protein-coding gene, O-linked N-acetylglucosamine transferase, which is overexpressed in epithelial cancer types. Consistent with the hypothesis that T-UCRs have important function in tumor formation, HINCUT-1 supports cell proliferation specifically under hypoxic conditions and may be critical for optimal O-GlcNAcylation of proteins when oxygen tension is limiting. Our data gives a first glimpse of a novel functional hypoxic network comprising protein-coding transcripts and noncoding RNAs (ncRNAs) from the T-UCRs category.

Fibroblast activation protein expression by stromal cells and tumor-associated macrophages in human breast cancer

Fibroblast activation protein (FAP) has long been known to be expressed in the stroma of breast cancer. However, very little is known if the magnitude of FAP expression within the stroma may have a prognostic value and reflect the heterogeneous biology of the tumor cell. An earlier study had suggested that stromal FAP expression in breast cancer was inversely proportional to prognosis. We, therefore, hypothesized that stromal FAP expression may correlate with clinicopathologic variables and may serve as an adjunct prognostic factor in breast cancer. We evaluated the expression of FAP in a panel of breast cancer tissues (n = 52) using a combination of immunostain analyses at the tissue and single-cell level using freshly frozen or freshly digested human breast tumor samples, respectively. Our results showed that FAP expression was abundantly expressed in the stroma across all breast cancer subtypes without significant correlation with clinicopathologic factors. We further identified a subset of FAP-positive (or FAP(+)) stromal cells that also expressed CD45, a pan-leukocyte marker. Using freshly dissociated human breast tumor specimens (n = 5), we demonstrated that some of these FAP(+)CD45(+) cells were CD11b(+)CD14(+)MHC-II(+), indicating that they were likely tumor-associated macrophages (TAMs). Although FAP(+)CD45(+) cells have been demonstrated in the mouse tumor stroma, our results demonstrating that human breast TAMs expressed FAP were novel and suggested that existing and future FAP-directed therapy may have dual-therapeutic benefits targeting both stromal mesenchymal cells and immune cells such as TAMs. More work is needed to explore the role of FAP as a potential targetable molecule in breast cancer treatment.

Isoform level expression profiles provide better cancer signatures than gene level expression profiles

BackgroundThe majority of mammalian genes generate multiple transcript variants and protein isoforms through alternative transcription and/or alternative splicing, and the dynamic changes at the transcript/isoform level between non-oncogenic and cancer cells remain largely unexplored. We hypothesized that isoform level expression profiles would be better than gene level expression profiles at discriminating between non-oncogenic and cancer cellsgene level.MethodsWe analyzed 160 Affymetrix exon-array datasets, comprising cell lines of non-oncogenic or oncogenic tissue origins. We obtained the transcript-level and gene level expression estimates, and used unsupervised and supervised clustering algorithms to study the profile similarity between the samples at both gene and isoform levels.ResultsHierarchical clustering, based on isoform level expressions, effectively grouped the non-oncogenic and oncogenic cell lines with a virtually perfect homogeneity-grouping rate (97.5%), regardless of the tissue origin of the cell lines. However, gene levelthis rate was much lower, being 75% at best based on the gene level expressions. Statistical analyses of the difference between cancer and non-oncogenic samples identified the existence of numerous genes with differentially expressed isoforms, which otherwise were not significant at the gene level. We also found that canonical pathways of protein ubiquitination, purine metabolism, and breast-cancer regulation by stathmin1 were significantly enriched among genes thatshow differential expression at isoform level but not at gene level.ConclusionsIn summary, cancer cell lines, regardless of their tissue of origin, can be effectively discriminated from non-cancer cell lines at isoform level, but not at gene level. This study suggests the existence of an isoform signature, rather than a gene signature, which could be used to distinguish cancer cells from normal cells.

Isoform-level gene signature improves prognostic stratification and accurately classifies glioblastoma subtypes

Molecular stratification of tumors is essential for developing personalized therapies. Although patient stratification strategies have been successful; computational methods to accurately translate the gene-signature from high-throughput platform to a clinically adaptable low-dimensional platform are currently lacking. Here, we describe PIGExClass (platform-independent isoform-level gene-expression based classification-system), a novel computational approach to derive and then transfer gene-signatures from one analytical platform to another. We applied PIGExClass to design a reverse transcriptase-quantitative polymerase chain reaction (RT-qPCR) based molecular-subtyping assay for glioblastoma multiforme (GBM), the most aggressive primary brain tumors. Unsupervised clustering of TCGA (the Cancer Genome Altas Consortium) GBM samples, based on isoform-level gene-expression profiles, recaptured the four known molecular subgroups but switched the subtype for 19% of the samples, resulting in significant (P = 0.0103) survival differences among the refined subgroups. PIGExClass derived four-class classifier, which requires only 121 transcript-variants, assigns GBM patients’ molecular subtype with 92% accuracy. This classifier was translated to an RT-qPCR assay and validated in an independent cohort of 206 GBM samples. Our results demonstrate the efficacy of PIGExClass in the design of clinically adaptable molecular subtyping assay and have implications for developing robust diagnostic assays for cancer patient stratification.

Global Identification of EVI1 Target Genes in Acute Myeloid Leukemia

The ecotropic virus integration site 1 (EVI1) transcription factor is associated with human myeloid malignancy of poor prognosis and is overexpressed in 8–10% of adult AML and strikingly up to 27% of pediatric MLL-rearranged leukemias. For the first time, we report comprehensive genomewide EVI1 binding and whole transcriptome gene deregulation in leukemic cells using a combination of ChIP-Seq and RNA-Seq expression profiling. We found disruption of terminal myeloid differentiation and cell cycle regulation to be prominent in EVI-induced leukemogenesis. Specifically, we identified EVI1 directly binds to and downregulates the master myeloid differentiation gene Cebpe and several of its downstream gene targets critical for terminal myeloid differentiation. We also found EVI1 binds to and downregulates Serpinb2 as well as numerous genes involved in the Jak-Stat signaling pathway. Finally, we identified decreased expression of several ATP-dependent P2X purinoreceptors genes involved in apoptosis mechanisms. These findings provide a foundation for future study of potential therapeutic gene targets for EVI1-induced leukemia.

NPEBseq: nonparametric empirical bayesian-based procedure for differential expression analysis of RNA-seq data

BackgroundRNA-seq, a massive parallel-sequencing-based transcriptome profiling method, provides digital data in the form of aligned sequence read counts. The comparative analyses of the data require appropriate statistical methods to estimate the differential expression of transcript variants across different cell/tissue types and disease conditions.ResultsWe developed a novel nonparametric empirical Bayesian-based approach (NPEBseq) to model the RNA-seq data. The prior distribution of the Bayesian model is empirically estimated from the data without any parametric assumption, and hence the method is “nonparametric” in nature. Based on this model, we proposed a method for detecting differentially expressed genes across different conditions. We also extended this method to detect differential usage of exons from RNA-seq data. The evaluation of NPEBseq on both simulated and publicly available RNA-seq datasets and comparison with three popular methods showed improved results for experiments with or without biological replicates.ConclusionsNPEBseq can successfully detect differential expression between different conditions not only at gene level but also at exon level from RNA-seq datasets. In addition, NPEBSeq performs significantly better than current methods and can be applied to genome-wide RNA-seq datasets. Sample datasets and R package are available at http://bioinformatics.wistar.upenn.edu/NPEBseq.

IsoformEx: isoform level gene expression estimation using weighted non-negative least squares from mRNA-Seq data

BackgroundmRNA-Seq technology has revolutionized the field of transcriptomics for identification and quantification of gene transcripts not only at gene level but also at isoform level. Estimating the expression levels of transcript isoforms from mRNA-Seq data is a challenging problem due to the presence of constitutive exons.ResultsWe propose a novel algorithm (IsoformEx) that employs weighted non-negative least squares estimation method to estimate the expression levels of transcript isoforms. Validations based on in silico simulation of mRNA-Seq and qRT-PCR experiments with real mRNA-Seq data showed that IsoformEx could accurately estimate transcript expression levels. In comparisons with published methods, the transcript expression levels estimated by IsoformEx showed higher correlation with known transcript expression levels from simulated mRNA-Seq data, and higher agreement with qRT-PCR measurements of specific transcripts for real mRNA-Seq data.ConclusionsIsoformEx is a fast and accurate algorithm to estimate transcript expression levels and gene expression levels, which takes into account short exons and alternative exons with a weighting scheme. The software is available at http://bioinformatics.wistar.upenn.edu/isoformex.

Tree-based position weight matrix approach to model transcription factor binding site profiles.

Most of the position weight matrix (PWM) based bioinformatics methods developed to predict transcription factor binding sites (TFBS) assume each nucleotide in the sequence motif contributes independently to the interaction between protein and DNA sequence, usually producing high false positive predictions. The increasing availability of TF enrichment profiles from recent ChIP-Seq methodology facilitates the investigation of dependent structure and accurate prediction of TFBSs. We develop a novel Tree-based PWM (TPWM) approach to accurately model the interaction between TF and its binding site. The whole tree-structured PWM could be considered as a mixture of different conditional-PWMs. We propose a discriminative approach, called TPD (TPWM based Discriminative Approach), to construct the TPWM from the ChIP-Seq data with a pre-existing PWM. To achieve the maximum discriminative power between the positive and negative datasets, the cutoff value is determined based on the Matthew Correlation Coefficient (MCC). The resulting TPWMs are evaluated with respect to accuracy on extensive synthetic datasets. We then apply our TPWM discriminative approach on several real ChIP-Seq datasets to refine the current TFBS models stored in the TRANSFAC database. Experiments on both the simulated and real ChIP-Seq data show that the proposed method starting from existing PWM has consistently better performance than existing tools in detecting the TFBSs. The improved accuracy is the result of modelling the complete dependent structure of the motifs and better prediction of true positive rate. The findings could lead to better understanding of the mechanisms of TF-DNA interactions.

Shift in Monocyte Apoptosis with Increasing Viral Load and Change in Apoptosis-Related ISG/Bcl2 Family Gene Expression in Chronically HIV-1-Infected Subjects

ABSTRACT Although monocytes and macrophages are targets of HIV-1-mediated immunopathology, the impact of high viremia on activation-induced monocyte apoptosis relative to monocyte and macrophage activation changes remains undetermined. In this study, we determined constitutive and oxidative stress-induced monocyte apoptosis in uninfected and HIV+ individuals across a spectrum of viral loads (n = 35; range, 2,243 to 1,355,998 HIV-1 RNA copies/ml) and CD4 counts (range, 26 to 801 cells/mm3). Both constitutive apoptosis and oxidative stress-induced apoptosis were positively associated with viral load and negatively associated with CD4, with an elevation in apoptosis occurring in patients with more than 40,000 (4.6 log) copies/ml. As expected, expression of Rb1 and interferon-stimulated genes (ISGs), plasma soluble CD163 (sCD163) concentration, and the proportion of CD14++ CD16+ intermediate monocytes were elevated in viremic patients compared to those in uninfected controls. Although CD14++ CD16+ frequencies, sCD14, sCD163, and most ISG expression were not directly associated with a change in apoptosis, sCD14 and ISG expression showed an association with increasing viral load. Multivariable analysis of clinical values and monocyte gene expression identified changes in IFI27, IFITM2, Rb1, and Bcl2 expression as determinants of constitutive apoptosis (P = 3.77 × 10−5; adjusted R 2 = 0.5983), while changes in viral load, IFITM2, Rb1, and Bax expression were determinants of oxidative stress-induced apoptosis (P = 5.59 × 10−5; adjusted R 2 = 0.5996). Our data demonstrate differential activation states in monocytes between levels of viremia in association with differences in apoptosis that may contribute to greater monocyte turnover with high viremia. IMPORTANCE This study characterized differential monocyte activation, apoptosis, and apoptosis-related gene expression in low- versus high-level viremic HIV-1 patients, suggesting a shift in apoptosis regulation that may be associated with disease state. Using single and multivariable analysis of monocyte activation parameters and gene expression, we supported the hypothesis that monocyte apoptosis in HIV disease is a reflection of viremia and activation state with contributions from gene expression changes within the ISG and Bcl2 gene families. Understanding monocyte apoptosis response may inform HIV immunopathogenesis, retention of infected macrophages, and monocyte turnover in low- or high-viral-load states.