Pediatric cardiomyopathies are a heterogeneous group of disorders in which early detection and treatment can drastically alter the course of the disease. Cardiac MRIs have become an increasingly popular option in evaluating for cardiomyopathies, though they remain expensive and time-consuming compared to echocardiography. Because guidelines for the use of cardiac MRI are vague, we investigated the quantitative characteristics of cardiac echo that predict subsequent positive cardiac MRI. Measurements were extracted from echo reports, processed, and fed into a Bayesian rule learning system. We discovered that ejection fraction and interventricular septum thickness were particularly important predictors of positive cardiac MRI. These features may help justify obtaining a cardiac MRI when echo results are inconclusive.

BACKGROUND: Most 'transcriptomic' data from microarrays are generated from small sample sizes compared to the large number of measured biomarkers, making it very difficult to build accurate and generalizable disease state classification models. Integrating information from different, but related, 'transcriptomic' data may help build better classification models. However, most proposed methods for integrative analysis of 'transcriptomic' data cannot incorporate domain knowledge, which can improve model performance. To this end, we have developed a methodology that leverages transfer rule learning and functional modules, which we call TRL-FM, to capture and abstract domain knowledge in the form of classification rules to facilitate integrative modeling of multiple gene expression data. TRL-FM is an extension of the transfer rule learner (TRL) that we developed previously. The goal of this study was to test our hypothesis that "an integrative model obtained via the TRL-FM approach outperforms traditional models based on single gene expression data sources". RESULTS: To evaluate the feasibility of the TRL-FM framework, we compared the area under the ROC curve (AUC) of models developed with TRL-FM and other traditional methods, using 21 microarray datasets generated from three studies on brain cancer, prostate cancer, and lung disease, respectively. The results show that TRL-FM statistically significantly outperforms TRL as well as traditional models based on single source data. In addition, TRL-FM performed better than other integrative models driven by meta-analysis and cross-platform data merging. CONCLUSIONS: The capability of utilizing transferred abstract knowledge derived from source data using feature mapping enables the TRL-FM framework to mimic the human process of learning and adaptation when performing related tasks. The novel TRL-FM methodology for integrative modeling for multiple 'transcriptomic' datasets is able to intelligently incorporate domain knowledge that traditional methods might disregard, to boost predictive power and generalization performance. In this study, TRL-FM's abstraction of knowledge is achieved in the form of functional modules, but the overall framework is generalizable in that different approaches of acquiring abstract knowledge can be integrated into this framework.

Mass-spectrometry (MS) based proteomics has become a key enabling technology for the systems approach to biology, providing insights into the protein complement of an organism. Bioinformatics analyses play a critical role in interpretation of large, and often replicated, MS datasets generated across laboratories and institutions. A significant amount of computational effort in the workflow is spent on the identification of protein and peptide components of complex biological samples, and consists of a series of steps relying on large database searches and intricate scoring algorithms. In this work, we share our efforts and experience in efficient handling of these large MS datasets through database indexing and parallelization based on multiprocessor architectures. We also identify important challenges and opportunities that are relevant specifically to the task of peptide and protein identification, and more generally to similar multi-step problems that are inherently parallelizable.

BACKGROUND: Several data mining methods require data that are discrete, and other methods often perform better with discrete data. We introduce an efficient Bayesian discretization (EBD) method for optimal discretization of variables that runs efficiently on high-dimensional biomedical datasets. The EBD method consists of two components, namely, a Bayesian score to evaluate discretizations and a dynamic programming search procedure to efficiently search the space of possible discretizations. We compared the performance of EBD to Fayyad and Irani's (FI) discretization method, which is commonly used for discretization. RESULTS: On 24 biomedical datasets obtained from high-throughput transcriptomic and proteomic studies, the classification performances of the C4.5 classifier and the naïve Bayes classifier were statistically significantly better when the predictor variables were discretized using EBD over FI. EBD was statistically significantly more stable to the variability of the datasets than FI. However, EBD was less robust, though not statistically significantly so, than FI and produced slightly more complex discretizations than FI. CONCLUSIONS: On a range of biomedical datasets, a Bayesian discretization method (EBD) yielded better classification performance and stability but was less robust than the widely used FI discretization method. The EBD discretization method is easy to implement, permits the incorporation of prior knowledge and belief, and is sufficiently fast for application to high-dimensional data.

We present a novel framework for integrative biomarker discovery from related but separate data sets created in biomarker profiling studies. The framework takes prior knowledge in the form of interpretable, modular rules, and uses them during the learning of rules on a new data set. The framework consists of two methods of transfer of knowledge from source to target data: transfer of whole rules and transfer of rule structures. We evaluated the methods on three pairs of data sets: one genomic and two proteomic. We used standard measures of classification performance and three novel measures of amount of transfer. Preliminary evaluation shows that whole-rule transfer improves classification performance over using the target data alone, especially when there is more source data than target data. It also improves performance over using the union of the data sets.

MOTIVATION: Disease state prediction from biomarker profiling studies is an important problem because more accurate classification models will potentially lead to the discovery of better, more discriminative markers. Data mining methods are routinely applied to such analyses of biomedical datasets generated from high-throughput 'omic' technologies applied to clinical samples from tissues or bodily fluids. Past work has demonstrated that rule models can be successfully applied to this problem, since they can produce understandable models that facilitate review of discriminative biomarkers by biomedical scientists. While many rule-based methods produce rules that make predictions under uncertainty, they typically do not quantify the uncertainty in the validity of the rule itself. This article describes an approach that uses a Bayesian score to evaluate rule models. RESULTS: We have combined the expressiveness of rules with the mathematical rigor of Bayesian networks (BNs) to develop and evaluate a Bayesian rule learning (BRL) system. This system utilizes a novel variant of the K2 algorithm for building BNs from the training data to provide probabilistic scores for IF-antecedent-THEN-consequent rules using heuristic best-first search. We then apply rule-based inference to evaluate the learned models during 10-fold cross-validation performed two times. The BRL system is evaluated on 24 published 'omic' datasets, and on average it performs on par or better than other readily available rule learning methods. Moreover, BRL produces models that contain on average 70% fewer variables, which means that the biomarker panels for disease prediction contain fewer markers for further verification and validation by bench scientists.

BACKGROUND: The incorporation of biological knowledge can enhance the analysis of biomedical data. We present a novel method that uses a proteomic knowledge base to enhance the performance of a rule-learning algorithm in identifying putative biomarkers of disease from high-dimensional proteomic mass spectral data. In particular, we use the Empirical Proteomics Ontology Knowledge Base (EPO-KB) that contains previously identified and validated proteomic biomarkers to select m/zs in a proteomic dataset prior to analysis to increase performance. RESULTS: We show that using EPO-KB as a pre-processing method, specifically selecting all biomarkers found only in the biofluid of the proteomic dataset, reduces the dimensionality by 95% and provides a statistically significantly greater increase in performance over no variable selection and random variable selection. CONCLUSION: Knowledge-based variable selection even with a sparsely-populated resource such as the EPO-KB increases overall performance of rule-learning for disease classification from high-dimensional proteomic mass spectra.

Protein secondary structure prediction is an important step towards understanding the relation between protein sequence and structure. However, most current prediction methods use features difficult for biologists to interpret. In this paper, we present a new method that applies information retrieval techniques to solve the problem:we extract a context sensitive biological vocabulary for protein sequences and apply text classification methods to predict protein secondary structure. Experimental results show that our method performs comparably to the state-of-art methods. Furthermore, the context sensitive vocabularies can serve as a useful tool to discover meaningful regular expression patterns for protein structures.

MOTIVATION: Protein secondary structure prediction is an important step towards understanding how proteins fold in three dimensions. Recent analysis by information theory indicates that the correlation between neighboring secondary structures are much stronger than that of neighboring amino acids. In this article, we focus on the combination problem for sequences, i.e. combining the scores or assignments from single or multiple prediction systems under the constraint of a whole sequence, as a target for improvement in protein secondary structure prediction. RESULTS: We apply several graphical chain models to solve the combination problem and show that they are consistently more effective than the traditional window-based methods. In particular, conditional random fields (CRFs) moderately improve the predictions for helices and, more importantly, for beta sheets, which are the major bottleneck for protein secondary structure prediction.

BACKGROUND: Conserved protein sequence motifs are short stretches of amino acid sequence patterns that potentially encode the function of proteins. Several sequence pattern searching algorithms and programs exist foridentifying candidate protein motifs at the whole genome level. However, a much needed and important task is to determine the functions of the newly identified protein motifs. The Gene Ontology (GO) project is an endeavor to annotate the function of genes or protein sequences with terms from a dynamic, controlled vocabulary and these annotations serve well as a knowledge base. RESULTS: This paper presents methods to mine the GO knowledge base and use the association between the GO terms assigned to a sequence and the motifs matched by the same sequence as evidence for predicting the functions of novel protein motifs automatically. The task of assigning GO terms to protein motifs is viewed as both a binary classification and information retrieval problem, where PROSITE motifs are used as samples for mode training and functional prediction. The mutual information of a motif and aGO term association is found to be a very useful feature. We take advantage of the known motifs to train a logistic regression classifier, which allows us to combine mutual information with other frequency-based features and obtain a probability of correct association. The trained logistic regression model has intuitively meaningful and logically plausible parameter values, and performs very well empirically according to our evaluation criteria. CONCLUSIONS: In this research, different methods for automatic annotation of protein motifs have been investigated. Empirical result demonstrated that the methods have a great potential for detecting and augmenting information about the functions of newly discovered candidate protein motifs.