Breast cancer is the most common malignancy in women and is responsible for hundreds of thousands of deaths annually. As with most cancers, it is a heterogeneous disease and different breast cancer subtypes are treated differently. Understanding the difference in prognosis for breast cancer based on its molecular and phenotypic features is one avenue for improving treatment by matching the proper treatment with molecular subtypes of the disease. In this work, we employed a competition-based approach to modeling breast cancer prognosis using large datasets containing genomic and clinical information and an online real-time leaderboard program used to speed feedback to the modeling team and to encourage each modeler to work towards achieving a higher ranked submission. We find that machine learning methods combined with molecular features selected based on expert prior knowledge can improve survival predictions compared to current best-in-class methodologies and that ensemble models trained across multiple user submissions systematically outperform individual models within the ensemble. We also find that model scores are highly consistent across multiple independent evaluations. This study serves as the pilot phase of a much larger competition open to the whole research community, with the goal of understanding general strategies for model optimization using clinical and molecular profiling data and providing an objective, transparent system for assessing prognostic models.

MOTIVATION: Copy number alterations (CNAs) associated with cancer are known to contribute to genomic instability and gene deregulation. Integrating CNAs with gene expression helps to elucidate the mechanisms by which CNAs act and to identify the transcriptional downstream targets of CNAs. Such analyses can help to sort functional driver events from the many accompanying passenger alterations. However, the way CNAs affect gene expression can vary in different cellular contexts, for example between different subtypes of the same cancer. Thus, it is important to develop computational approaches capable of inferring differential connectivity of regulatory networks in different cellular contexts. RESULTS: We propose a statistical deregulation model that integrates copy number and expression data of different disease subtypes to jointly model common and differential regulatory relationships. Our model not only identifies CNAs driving gene expression changes, but at the same time also predicts differences in regulation that distinguish one cancer subtype from the other. We implement our model in a penalized regression framework and demonstrate in a simulation study the feasibility and accuracy of our approach. Subsequently, we show that this model can identify both known and novel aspects of cross-talk between the ER and NOTCH pathways in ER-negative-specific deregulations, when compared with ER-positive breast cancer. This flexible model can be applied on other modalities such as methylation or microRNA and expression to disentangle cancer signaling pathways. AVAILABILITY: The Bioconductor-compliant R package DANCE is available from www.markowetzlab.org/software/ CONTACT: [email protected]; [email protected].

SUMMARY: Several methods have been proposed to detect copy number changes and recurrent regions of copy number variation from aCGH, but few methods return probabilities of alteration explicitly, which are the direct answer to the question 'is this probe/region altered?' RJaCGH fits a Non-Homogeneous Hidden Markov model to the aCGH data using Markov Chain Monte Carlo with Reversible Jump, and returns the probability that each probe is gained or lost. Using these probabilites, recurrent regions (over sets of individuals) of copy number alteration can be found. AVAILABILITY: RJaCGH is available as an R package from CRAN repositories (e.g. http://cran.r-project.org/web/packages).

BACKGROUND: Alterations in the number of copies of genomic DNA that are common or recurrent among diseased individuals are likely to contain disease-critical genes. Unfortunately, defining common or recurrent copy number alteration (CNA) regions remains a challenge. Moreover, the heterogeneous nature of many diseases requires that we search for common or recurrent CNA regions that affect only some subsets of the samples (without knowledge of the regions and subsets affected), but this is neglected by most methods. RESULTS: We have developed two methods to define recurrent CNA regions from aCGH data. Our methods are unique and qualitatively different from existing approaches: they detect regions over both the complete set of arrays and alterations that are common only to some subsets of the samples (i.e., alterations that might characterize previously unknown groups); they use probabilities of alteration as input and return probabilities of being a common region, thus allowing researchers to modify thresholds as needed; the two parameters of the methods have an immediate, straightforward, biological interpretation. Using data from previous studies, we show that we can detect patterns that other methods miss and that researchers can modify, as needed, thresholds of immediate interpretability and develop custom statistics to answer specific research questions. CONCLUSION: These methods represent a qualitative advance in the location of recurrent CNA regions, highlight the relevance of population heterogeneity for definitions of recurrence, and can facilitate the clustering of samples with respect to patterns of CNA. Ultimately, the methods developed can become important tools in the search for genomic regions harboring disease-critical genes.

BACKGROUND: Yu et al. (BMC Bioinformatics 2007,8: 145+) have recently compared the performance of several methods for the detection of genomic amplification and deletion breakpoints using data from high-density single nucleotide polymorphism arrays. One of the methods compared is our non-homogenous Hidden Markov Model approach. Our approach uses Markov Chain Monte Carlo for inference, but Yu et al. ran the sampler for a severely insufficient number of iterations for a Markov Chain Monte Carlo-based method. Moreover, they did not use the appropriate reference level for the non-altered state. METHODS: We rerun the analysis in Yu et al. using appropriate settings for both the Markov Chain Monte Carlo iterations and the reference level. Additionally, to show how easy it is to obtain answers to additional specific questions, we have added a new analysis targeted specifically to the detection of breakpoints. RESULTS: The reanalysis shows that the performance of our method is comparable to that of the other methods analyzed. In addition, we can provide probabilities of a given spot being a breakpoint, something unique among the methods examined. CONCLUSION: Markov Chain Monte Carlo methods require using a sufficient number of iterations before they can be assumed to yield samples from the distribution of interest. Running our method with too small a number of iterations cannot be representative of its performance. Moreover, our analysis shows how our original approach can be easily adapted to answer specific additional questions (e.g., identify edges).

Genomic DNA copy-number alterations (CNAs) are associated with complex diseases, including cancer: CNAs are indeed related to tumoral grade, metastasis, and patient survival. CNAs discovered from array-based comparative genomic hybridization (aCGH) data have been instrumental in identifying disease-related genes and potential therapeutic targets. To be immediately useful in both clinical and basic research scenarios, aCGH data analysis requires accurate methods that do not impose unrealistic biological assumptions and that provide direct answers to the key question, "What is the probability that this gene/region has CNAs?" Current approaches fail, however, to meet these requirements. Here, we introduce reversible jump aCGH (RJaCGH), a new method for identifying CNAs from aCGH; we use a nonhomogeneous hidden Markov model fitted via reversible jump Markov chain Monte Carlo; and we incorporate model uncertainty through Bayesian model averaging. RJaCGH provides an estimate of the probability that a gene/region has CNAs while incorporating interprobe distance and the capability to analyze data on a chromosome or genome-wide basis. RJaCGH outperforms alternative methods, and the performance difference is even larger with noisy data and highly variable interprobe distance, both commonly found features in aCGH data. Furthermore, our probabilistic method allows us to identify minimal common regions of CNAs among samples and can be extended to incorporate expression data. In summary, we provide a rigorous statistical framework for locating genes and chromosomal regions with CNAs with potential applications to cancer and other complex human diseases.