Renato Vimieiro

Impact of androgen suppression and zoledronic acid on bone mineral density and fractures in the Trans-Tasman Radiation Oncology Group (TROG) 03.04 Randomised Androgen Deprivation and Radiotherapy (RADAR) randomized controlled trial for locally advanced prostate cancer.

To study the influence of adjuvant androgen suppression and bisphosphonates on incident vertebral and non‐spinal fracture rates and bone mineral density (BMD) in men with locally advanced prostate cancer.

The Discovery of Novel Biomarkers Improves Breast Cancer Intrinsic Subtype Prediction and Reconciles the Labels in the METABRIC Data Set

Background The prediction of breast cancer intrinsic subtypes has been introduced as a valuable strategy to determine patient diagnosis and prognosis, and therapy response. The PAM50 method, based on the expression levels of 50 genes, uses a single sample predictor model to assign subtype labels to samples. Intrinsic errors reported within this assay demonstrate the challenge of identifying and understanding the breast cancer groups. In this study, we aim to: a) identify novel biomarkers for subtype individuation by exploring the competence of a newly proposed method named CM1 score, and b) apply an ensemble learning, as opposed to the use of a single classifier, for sample subtype assignment. The overarching objective is to improve class prediction. Methods and Findings The microarray transcriptome data sets used in this study are: the METABRIC breast cancer data recorded for over 2000 patients, and the public integrated source from ROCK database with 1570 samples. We first computed the CM1 score to identify the probes with highly discriminative patterns of expression across samples of each intrinsic subtype. We further assessed the ability of 42 selected probes on assigning correct subtype labels using 24 different classifiers from the Weka software suite. For comparison, the same method was applied on the list of 50 genes from the PAM50 method. Conclusions The CM1 score portrayed 30 novel biomarkers for predicting breast cancer subtypes, with the confirmation of the role of 12 well-established genes. Intrinsic subtypes assigned using the CM1 list and the ensemble of classifiers are more consistent and homogeneous than the original PAM50 labels. The new subtypes show accurate distributions of current clinical markers ER, PR and HER2, and survival curves in the METABRIC and ROCK data sets. Remarkably, the paradoxical attribution of the original labels reinforces the limitations of employing a single sample classifiers to predict breast cancer intrinsic subtypes.

A new method for mining disjunctive emerging patterns in high-dimensional datasets using hypergraphs

We investigate in this paper the problem of mining disjunctive emerging patterns in high-dimensional biomedical datasets. Disjunctive emerging patterns are sets of features that are very frequent among samples of a target class, cases in a case-control study, for example, and are very rare among all other samples. We, for the very first time, demonstrate that this problem can be solved using minimal transversals in a hypergraph. We propose a new divide-and-conquer algorithm that enables us to efficiently compute disjunctive emerging patterns in parallel and distributed environments. We conducted experiments using real-world microarray gene expression datasets to assess the performance of our approach. Our results show that our approach is more efficient than the state-of-the-art solution available in the literature. In this sense, we contribute to the area of bioinformatics and data mining by providing another useful alternative to identify patterns distinguishing samples with different class labels, such as those in case-control studies, for example.

Disclosed: An efficient depth-first, top-down algorithm for mining disjunctive closed itemsets in high-dimensional data

Abstract We focus, in this paper, on the computational challenges of identifying disjunctive Boolean patterns in high-dimensional data. We conduct our analysis focusing particularly in microarray gene expression data, since this is one of the most stereotypical examples of high-dimensional data. We devised a novel algorithm that takes advantage of the scarcity of samples in microarray data sets, allowing us to efficiently find disjunctive closed patterns. Our algorithm, Disclosed , mines disjunctive closed itemsets by exploring the search space in a depth-first, top-down manner. We evaluated the performance of our algorithm to execute such a task using real microarray gene expression data sets publicly available on the Internet. Our experiments revealed under what situations, the characteristics of a data set, our method obtain a good , bad or average performance. We also compared the performance of our method with the state of the art algorithms for finding disjunctive closed patterns and disjunctive minimal generators. We observed that our approach is two orders of magnitude more efficient, both in terms of time and memory.

Mining disjunctive minimal generators with TitanicOR

Disjunctive minimal generators were proposed by Zhao, Zaki, and Ramakrishnan (2006). They defined disjunctive closed itemsets and disjunctive minimal generators through the disjunctive support function. We prove that the disjunctive support function is compatible with the closure operator presented by Zhao et al. (2006). Such compatibility allows us to adapt the original version of the Titanic algorithm, proposed by Stumme, Taouil, Bastide, Pasquier, and Lakhal (2002) to mine iceberg concept lattices and closed itemsets, to mine disjunctive minimal generators. We present TitanicOR, a new breadth-first algorithm for mining disjunctive minimal generators. We evaluate the performance of our method with both synthetic and real data sets and compare TitanicORs performance with the performance of BLOSOM (Zhao et al., 2006), the state of the art method and sole algorithm available prior to TitanicOR for mining disjunctive minimal generators. We show that TitanicORs breadth-first approach is up to two orders of magnitude faster than BLOSOMs depth-first approach.

Iteratively refining breast cancer intrinsic subtypes in the METABRIC dataset

BackgroundMulti-gene lists and single sample predictor models have been currently used to reduce the multidimensional complexity of breast cancers, and to identify intrinsic subtypes. The perceived inability of some models to deal with the challenges of processing high-dimensional data, however, limits the accurate characterisation of these subtypes. Towards the development of robust strategies, we designed an iterative approach to consistently discriminate intrinsic subtypes and improve class prediction in the METABRIC dataset.FindingsIn this study, we employed the CM1 score to identify the most discriminative probes for each group, and an ensemble learning technique to assess the ability of these probes on assigning subtype labels using 24 different classifiers. Our analysis is comprised of an iterative computation of these methods and statistical measures performed on a set of over 2000 samples. The refined labels assigned using this iterative approach revealed to be more consistent and in better agreement with clinicopathological markers and patients’ overall survival than those originally provided by the PAM50 method.ConclusionsThe assignment of intrinsic subtypes has a significant impact in translational research for both understanding and managing breast cancer. The refined labelling, therefore, provides more accurate and reliable information by improving the source of fundamental science prior to clinical applications in medicine.

Identification of Genome-Wide SNP–SNP and SNP–Clinical Boolean Interactions in Age-Related Macular Degeneration

We propose here a methodology to uncover modularities in the network of SNP-SNP interactions most associated with disease. We start by computing all possible Boolean binary SNP interactions across the whole genome. By constructing a weighted graph of the most relevant interactions and via a combinatorial optimization approach, we find the most highly interconnected SNPs. We show that the method can be easily extended to find SNP/environment interactions. Using a modestly sized GWAS dataset of age-related macular degeneration (AMD), we identify a group of only 19 SNPs, which include those in previously reported regions associated to AMD. We also uncover a larger set of loci pointing to a matrix of key processes and functions that are affected. The proposed integrative methodology extends and overlaps traditional statistical analysis in a natural way. Combinatorial optimization techniques allow us to find the kernel of the most central interactions, complementing current methods of GWAS analysis and also enhancing the search for gene-environment interaction.

SSDP: A Simple Evolutionary Approach for Top-K Discriminative Patterns in High Dimensional Databases

It is a great challenge to companies, governments and researchers to extract knowledge in high dimensional databases. Discriminative Patterns (DPs) is an area of data mining that aims to extract relevant and readable information in databases with target attribute. Among the algorithms developed for search DPs, it has highlighted the use of evolutionary computing. However, the evolutionary approaches typically (1) are not adapted for high dimensional problems and (2) have many nontrivial parameters. This paper presents SSDP (Simple Search Discriminative Patterns), an evolutionary approach to search the top-k DPs adapted to high dimensional databases that use only two easily adjustable external parameters.