Rafael Carreira

BioDR: Semantic indexing networks for biomedical document retrieval

In Biomedical research, retrieving documents that match an interesting query is a task performed quite frequently. Typically, the set of obtained results is extensive containing many non-interesting documents and consists in a flat list, i.e., not organized or indexed in any way. This work proposes BioDR, a novel approach that allows the semantic indexing of the results of a query, by identifying relevant terms in the documents. These terms emerge from a process of Named Entity Recognition that annotates occurrences of biological terms (e.g. genes or proteins) in abstracts or full-texts. The system is based on a learning process that builds an Enhanced Instance Retrieval Network (EIRN) from a set of manually classified documents, regarding their relevance to a given problem. The resulting EIRN implements the semantic indexing of documents and terms, allowing for enhanced navigation and visualization tools, as well as the assessment of relevance for new documents.

Metabolic Profiling and Classification of Propolis Samples from Southern Brazil: An NMR-Based Platform Coupled with Machine Learning

The chemical composition of propolis is affected by environmental factors and harvest season, making it difficult to standardize its extracts for medicinal usage. By detecting a typical chemical profile associated with propolis from a specific production region or season, certain types of propolis may be used to obtain a specific pharmacological activity. In this study, propolis from three agroecological regions (plain, plateau, and highlands) from southern Brazil, collected over the four seasons of 2010, were investigated through a novel NMR-based metabolomics data analysis workflow. Chemometrics and machine learning algorithms (PLS-DA and RF), including methods to estimate variable importance in classification, were used in this study. The machine learning and feature selection methods permitted construction of models for propolis sample classification with high accuracy (>75%, reaching ∼90% in the best case), better discriminating samples regarding their collection seasons comparatively to the harvest regions. PLS-DA and RF allowed the identification of biomarkers for sample discrimination, expanding the set of discriminating features and adding relevant information for the identification of the class-determining metabolites. The NMR-based metabolomics analytical platform, coupled to bioinformatic tools, allowed characterization and classification of Brazilian propolis samples regarding the metabolite signature of important compounds, i.e., chemical fingerprint, harvest seasons, and production regions.

Semantic annotation of biological concepts interplaying microbial cellular responses

BackgroundAutomated extraction systems have become a time saving necessity in Systems Biology. Considerable human effort is needed to model, analyse and simulate biological networks. Thus, one of the challenges posed to Biomedical Text Mining tools is that of learning to recognise a wide variety of biological concepts with different functional roles to assist in these processes.ResultsHere, we present a novel corpus concerning the integrated cellular responses to nutrient starvation in the model-organism Escherichia coli. Our corpus is a unique resource in that it annotates biomedical concepts that play a functional role in expression, regulation and metabolism. Namely, it includes annotations for genetic information carriers (genes and DNA, RNA molecules), proteins (transcription factors, enzymes and transporters), small metabolites, physiological states and laboratory techniques. The corpus consists of 130 full-text papers with a total of 59043 annotations for 3649 different biomedical concepts; the two dominant classes are genes (highest number of unique concepts) and compounds (most frequently annotated concepts), whereas other important cellular concepts such as proteins account for no more than 10% of the annotated concepts.ConclusionsTo the best of our knowledge, a corpus that details such a wide range of biological concepts has never been presented to the text mining community. The inter-annotator agreement statistics provide evidence of the importance of a consolidated background when dealing with such complex descriptions, the ambiguities naturally arising from the terminology and their impact for modelling purposes.Availability is granted for the full-text corpora of 130 freely accessible documents, the annotation scheme and the annotation guidelines. Also, we include a corpus of 340 abstracts.

A machine learning and chemometrics assisted interpretation of spectroscopic data --- a NMR-Based metabolomics platform for the assessment of brazilian propolis

In this work, a metabolomics dataset from 1H nuclear magnetic resonance spectroscopy of Brazilian propolis was analyzed using machine learning algorithms, including feature selection and classification methods. Partial least square-discriminant analysis (PLS-DA), random forest (RF), and wrapper methods combining decision trees and rules with evolutionary algorithms (EA) showed to be complementary approaches, allowing to obtain relevant information as to the importance of a given set of features, mostly related to the structural fingerprint of aliphatic and aromatic compounds typically found in propolis, e.g., fatty acids and phenolic compounds. The feature selection and decision tree-based algorithms used appear to be suitable tools for building classification models for the Brazilian propolis metabolomics regarding its geographic origin, with consistency, high accuracy, and avoiding redundant information as to the metabolic signature of relevant compounds.

CBFA: phenotype prediction integrating metabolic models with constraints derived from experimental data

BackgroundFlux analysis methods lie at the core of Metabolic Engineering (ME), providing methods for phenotype simulation that allow the determination of flux distributions under different conditions. Although many constraint-based modeling software tools have been developed and published, none provides a free user-friendly application that makes available the full portfolio of flux analysis methods.ResultsThis work presents Constraint-based Flux Analysis (CBFA), an open-source software application for flux analysis in metabolic models that implements several methods for phenotype prediction, allowing users to define constraints associated with measured fluxes and/or flux ratios, together with environmental conditions (e.g. media) and reaction/gene knockouts. CBFA identifies the set of applicable methods based on the constraints defined from user inputs, encompassing algebraic and constraint-based simulation methods. The integration of CBFA within the OptFlux framework for ME enables the utilization of different model formats and standards and the integration with complementary methods for phenotype simulation and visualization of results.ConclusionsA general-purpose and flexible application is proposed that is independent of the origin of the constraints defined for a given simulation. The aim is to provide a simple to use software tool focused on the application of several flux prediction methods.

Genetic variants as ovarian cancer first-line treatment hallmarks: A systematic review and meta-analysis

BACKGROUND The potential predictive value of genetic polymorphisms in ovarian cancer first-line treatment is inconsistently reported. We aimed to review ovarian cancer pharmacogenetic studies to update and summarize the available data and to provide directions for further research. METHODS A systematic review followed by a meta-analysis was conducted on cohort studies assessing the involvement of genetic polymorphisms in ovarian cancer first-line treatment response retrieved through a MEDLINE database search by November 2016. Studies were pooled and summary estimates and 95% confidence intervals (CI) were calculated using random or fixed-effects models as appropriate. RESULTS One hundred and forty-two studies gathering 106871 patients were included. Combined data suggested that GSTM1-null genotype patients have a lower risk of death compared to GSTM1-wt carriers, specifically in advanced stages (hazard ratio (HR), 0.68; 95% CI, 0.48-0.97) and when submitted to platinum-based chemotherapy (aHR, 0.61; 95% CI, 0.39-0.94). ERCC1 rs11615 and rs3212886 might have also a significant impact in treatment outcome (aHR, 0.67; 95% CI, 0.51-0.89; aHR, 1.28; 95% CI, 1.01-1.63, respectively). Moreover, ERCC2 rs13181 and rs1799793 showed a distinct ethnic behavior (Asians: aHR, 1.41; 95% CI, 0.80-2.49; aHR, 1.07; 95% CI, 0.62-1.86; Caucasians: aHR, 0.10; 95% CI, 0.01-0.96; aHR, 0.18; 95% CI, 0.05-0.68, respectively). CONCLUSION(S) The definition of integrative predictive models should encompass genetic information, especially regarding GSTM1 homozygous deletion. Justifying additional pharmacogenetic investigation are variants in ERCC1 and ERCC2, which highlight the DNA Repair ability to ovarian cancer prognosis. Further knowledge could aid to understand platinum-treatment failure and to tailor chemotherapy strategies.

Biomedical Text Mining Applied to Document Retrieval and Semantic Indexing

In Biomedical research, the ability to retrieve the adequate information from the ever growing literature is an extremely important asset. This work provides an enhanced and general purpose approach to the process of document retrieval that enables the filtering of PubMed query results. The system is based on semantic indexing providing, for each set of retrieved documents, a network that links documents and relevant terms obtained by the annotation of biological entities (e.g. genes or proteins). This network provides distinct user perspectives and allows navigation over documents with similar terms and is also used to assess document relevance. A network learning procedure, based on previous work from e-mail spam filtering, is proposed, receiving as input a training set of manually classified documents.

A framework for the development of Biomedical Text Mining software tools

Over the last few years, a growing number of techniques has been successfully proposed to tackle diverse challenges in the biomedical text mining (BioTM) arena. However, the set of available software tools to researchers has not grown in a similar way. This work makes a contribution to close this gap, proposing a framework to ease the development of user-friendly and interoperable applications in this field, based on a set of available modular components. These modules can be connected in diverse ways to create applications that fit distinct user roles. Also, developers of new algorithms have a framework that allows them to easily integrate their implementations with state-of-the-art BioTM software for related tasks.

Algorithms to infer metabolic flux ratios from fluxomics data

In silico cell simulation approaches based in the use of genome-scale metabolic models (GSMMs) and constraint-based methods such as Flux Balance Analysis are gaining importance, but methods to integrate these approaches with omics data are still greatly needed. In this work, the focus relies on fluxomics data that provide valuable information on the intracellular fluxes, although in many cases in an indirect, incomplete and noisy way. The proposed framework enables the integration of fluxomics data, in the form of 13C labeling distribution for metabolite fragments, with GSMMs enriched with carbon atom transition maps. The algorithms implemented allow to infer labeling distributions for fragments/metabolites not measured and to build expressions for the relevant flux ratios that can be then used to enrich constraint-based methods for flux determination. This approach does not require any assumptions on the metabolic network and reaction reversibility, allowing to compute ratios originating from coupled joint points of the network. Also, when enough data do not exist, the system tries to infer ratio bounds from the measurements.