Single-cell genetic analysis reveals insights into clonal development of prostate cancers and indicates loss of PTEN as a marker of poor prognosis.

Gauging the risk of developing progressive disease is a major challenge in prostate cancer patient management. We used genetic markers to understand genomic alteration dynamics during disease progression. By using a novel, advanced, multicolor fluorescence in situ hybridization approach, we enumerated copy numbers of six genes previously identified by array comparative genomic hybridization to be involved in aggressive prostate cancer [TBL1XR1, CTTNBP2, MYC (alias c-myc), PTEN, MEN1, and PDGFB] in six nonrecurrent and seven recurrent radical prostatectomy cases. An ERG break-apart probe to detect TMPRSS2-ERG fusions was included. Subsequent hybridization of probe panels and cell relocation resulted in signal counts for all probes in each individual cell analyzed. Differences in the degree of chromosomal and genomic instability (ie, tumor heterogeneity) or the percentage of cells with TMPRSS2-ERG fusion between samples with or without progression were not observed. Tumors from patients that progressed had more chromosomal gains and losses, and showed a higher degree of selection for a predominant clonal pattern. PTEN loss was the most frequent aberration in progressers (57%), followed by TBL1XR1 gain (29%). MYC gain was observed in one progresser, which was the only lesion with an ERG gain, but no TMPRSS2-ERG fusion. According to our results, a probe set consisting of PTEN, MYC, and TBL1XR1 would detect progressers with 86% sensitivity and 100% specificity. This will be evaluated further in larger studies.

Network signatures of survival in glioblastoma multiforme.

To determine a molecular basis for prognostic differences in glioblastoma multiforme (GBM), we employed a combinatorial network analysis framework to exhaustively search for molecular patterns in protein-protein interaction (PPI) networks. We identified a dysregulated molecular signature distinguishing short-term (survival<225 days) from long-term (survival>635 days) survivors of GBM using whole genome expression data from The Cancer Genome Atlas (TCGA). A 50-gene subnetwork signature achieved 80% prediction accuracy when tested against an independent gene expression dataset. Functional annotations for the subnetwork signature included "protein kinase cascade," "IκB kinase/NFκB cascade," and "regulation of programmed cell death" - all of which were not significant in signatures of existing subtypes. Finally, we used label-free proteomics to examine how our subnetwork signature predicted protein level expression differences in an independent GBM cohort of 16 patients. We found that the genes discovered using network biology had a higher probability of dysregulated protein expression than either genes exhibiting individual differential expression or genes derived from known GBM subtypes. In particular, the long-term survivor subtype was characterized by increased protein expression of DNM1 and MAPK1 and decreased expression of HSPA9, PSMD3, and CANX. Overall, we demonstrate that the combinatorial analysis of gene expression data constrained by PPIs outlines an approach for the discovery of robust and translatable molecular signatures in GBM.

Single-cell genetic analysis of ductal carcinoma in situ and invasive breast cancer reveals enormous tumor heterogeneity yet conserved genomic imbalances and gain of MYC during progression.

Ductal carcinoma in situ (DCIS) is a precursor lesion of invasive ductal carcinoma (IDC) of the breast. To understand the dynamics of genomic alterations in this progression, we used four multicolor fluorescence in situ hybridization probe panels consisting of the oncogenes COX2, MYC, HER2, CCND1, and ZNF217 and the tumor suppressor genes DBC2, CDH1, and TP53 to visualize copy number changes in 13 cases of synchronous DCIS and IDC based on single-cell analyses. The DCIS had a lower degree of chromosomal instability than the IDC. Despite enormous intercellular heterogeneity in DCIS and IDC, we observed signal patterns consistent with a nonrandom distribution of genomic imbalances. CDH1 was most commonly lost, and gain of MYC emerged during progression from DCIS to IDC. Four of 13 DCISs showed identical clonal imbalances in the IDCs. Six cases revealed a switch, and in four of those, the IDC had acquired a gain of MYC. In one case, the major clone in the IDC was one of several clones in the DCIS, and in another case, the major clone in the DCIS became one of the two major clones in the IDC. Despite considerable chromosomal instability, in most cases the evolution from DCIS to IDC is determined by recurrent patterns of genomic imbalances, consistent with a biological continuum.

Identifying stage-specific protein subnetworks for colorectal cancer.

BACKGROUND: In recent years, many algorithms have been developed for network-based analysis of differential gene expression in complex diseases. These algorithms use protein-protein interaction (PPI) networks as an integrative framework and identify subnetworks that are coordinately dysregulated in the phenotype of interest. MOTIVATION: While such dysregulated subnetworks have demonstrated significant improvement over individual gene markers for classifying phenotype, the current state-of-the-art in dysregulated subnetwork discovery is almost exclusively limited to binary phenotype classes. However, many clinical applications require identification of molecular markers for multiple classes. APPROACH: We consider the problem of discovering groups of genes whose expression signatures can discriminate multiple phenotype classes. We consider two alternate formulations of this problem (i) an all-vs-all approach that aims to discover subnetworks distinguishing all classes, (ii) a one-vs-all approach that aims to discover subnetworks distinguishing each class from the rest of the classes. For the one-vs-all formulation, we develop a set-cover based algorithm, which aims to identify groups of genes such that at least one gene in the group exhibits differential expression in the target class. RESULTS: We test the proposed algorithms in the context of predicting stages of colorectal cancer. Our results show that the set-cover based algorithm identifying "stage-specific" subnetworks outperforms the all-vs-all approaches in classification. We also investigate the merits of utilizing PPI networks in the search for multiple markers, and show that, with correct parameter settings, network-guided search improves performance. Furthermore, we show that assessing statistical significance when selecting features greatly improves classification performance.

Protein-protein interaction networks and subnetworks in the biology of disease.

The main goal of systems medicine is to provide predictive models of the patho-physiology of complex diseases as well as define healthy states. The reason is clear--we hope accurate models will ultimately lead to more specific and sensitive markers of disease that will help clinicians better stratify their patient populations and optimize treatment plans. In addition, we expect that these models will define novel targets for combating disease. However, for many complex diseases, particularly at the clinical level, it is becoming increasingly clear that one or a few genomic variations alone (e.g., simple models) cannot adequately explain the multiple phenotypes related to disease states, or the variable risks that attend disease progression. We suggest that models that account for the activities of many interacting proteins will explain a wider range of variability inherent in these phenotypes. These models, which encompass protein interaction networks dysregulated for specific diseases and specific patient sub-populations, will be constructed by integrating protein interaction data with multiple types of other relevant cellular information. Protein interaction databases are thus playing an increasingly important role in systems biology approaches to the study of disease. They present us with a static, but highly functional view of the cellular state, and thus give us a better understanding of not only the normal phenotype, but also the overall disease phenotype at the level of the whole organism when certain interactions become dysregulated.

Subnetwork state functions define dysregulated subnetworks in cancer.

Emerging research demonstrates the potential of protein-protein interaction (PPI) networks in uncovering the mechanistic bases of cancers, through identification of interacting proteins that are coordinately dysregulated in tumorigenic and metastatic samples. When used as features for classification, such coordinately dysregulated subnetworks improve diagnosis and prognosis of cancer considerably over single-gene markers. However, existing methods formulate coordination between multiple genes through additive representation of their expression profiles and utilize fast heuristics to identify dysregulated subnetworks, which may not be well suited to the potentially combinatorial nature of coordinate dysregulation. Here, we propose a combinatorial formulation of coordinate dysregulation and decompose the resulting objective function to cast the problem as one of identifying subnetwork state functions that are indicative of phenotype. Based on this formulation, we show that coordinate dysregulation of larger subnetworks can be bounded using simple statistics on smaller subnetworks. We then use these bounds to devise an efficient algorithm, Crane, that can search the subnetwork space more effectively than existing algorithms. Comprehensive cross-classification experiments show that subnetworks identified by Crane outperform those identified by additive algorithms in predicting metastasis of colorectal cancer (CRC).

Systems biology analyses of gene expression and genome wide association study data in obstructive sleep apnea.

The precise molecular etiology of obstructive sleep apnea (OSA) is unknown; however recent research indicates that several interconnected aberrant pathways and molecular abnormalities are contributors to OSA. Identifying the genes and pathways associated with OSA can help to expand our understanding of the risk factors for the disease as well as provide new avenues for potential treatment. Towards these goals, we have integrated relevant high dimensional data from various sources, such as genome-wide expression data (microarray), protein-protein interaction (PPI) data and results from genome-wide association studies (GWAS) in order to define sub-network elements that connect some of the known pathways related to the disease as well as define novel regulatory modules related to OSA. Two distinct approaches are applied to identify sub-networks significantly associated with OSA. In the first case we used a biased approach based on sixty genes/proteins with known associations with sleep disorders and/or metabolic disease to seed a search using commercial software to discover networks associated with disease followed by information theoretic (mutual information) scoring of the sub-networks. In the second case we used an unbiased approach and generated an interactome constructed from publicly available gene expression profiles and PPI databases, followed by scoring of the network with p-values from GWAS data derived from OSA patients to uncover sub-networks significant for the disease phenotype. A comparison of the approaches reveals a number of proteins that have been previously known to be associated with OSA or sleep. In addition, our results indicate a novel association of Phosphoinositide 3-kinase, the STAT family of proteins and its related pathways with OSA.

Identification of coordinately dysregulated subnetworks in complex phenotypes.

In the study of complex phenotypes, single gene markers can only provide limited insights into the manifestation of phenotype. To this end, protein-protein interaction (PPI) networks prove useful in the identification of multiple interacting markers. Recent studies show that, when considered together, many proteins that are connected via physical and functional interactions exhibit significant differential expression with respect to various complex phenotypes, including cancers. As compared to single gene markers, these "coordinately dysregulated subnetworks" improve diagnosis and prognosis of cancer significantly and offer novel insights into the network dynamics of phenotype. However, the problem of identifying coordinately dysregulated subnetworks presents significant algorithmic challenges. Existing approaches utilize heuristics that aim to greedily maximize information-theoretic class separability measures, however, by definition of "coordinate" dysregulation, such greedy algorithms do not suit well to this problem. In this paper, we formulate coordinate dysregulation in the context of the well-known set-cover problem, with a view to capturing the coordination between multiple genes at a sample-specific resolution. Based on this formulation, we adapt state-of-the-art approximation algorithms for set-cover to the identification of coordinately dysregulated subnetworks. Comprehensive experimental results on human colorectal cancer (CRC) show that, when compared to existing algorithms, the proposed algorithm, NETCOVER, improves diagnosis of cancer and prediction of metastasis significantly. Our results also demonstrate that subnetworks in the neighborhood of known CRC driver genes exhibit significant coordinate dysregulation, indicating that the notion of coordinate dysregulation may indeed be useful in understanding the network dynamics of complex phenotypes.