Giuseppe Tradigo

SIGMCC: A system for sharing meta patient records in a Peer-to-Peer environment

This paper considers the interoperability and information sharing between health care providers. It proposes a distributed Peer-to-Peer (P2P) based framework that enables health operators of different hospitals to share and aggregate clinical information about patients and therapy effects. Patient records are mapped into a simple XML-based meta-Electronic Patient Record (meta-EPR). The meta-EPR is not a standard EPR proposal, but it is a lightweight data structure defined to contain relevant and aggregate information extracted from the different EPRs adopted by each hospital. Hospital operators formulate queries against meta-EPR schema; queries are then distributed to the connected hospitals hosting meta-EPR instances, through a P2P infrastructure. The presented framework has been fully implemented in a system called SIGMCC, which offers an Application Programming Interface (API) for query formulation, data loading and updating. As a case study, an application of the proposed meta-EPR to the cancer medical domain has been developed. Finally, SIGMCC implements a view mechanism to allow personal (patient) information protection against unauthorized users.

Using ontologies for preprocessing and mining spectra data on the grid

The analysis of mass spectrometry proteomics data requires the composition of different software tools devoted to the loading, management, preprocessing, mining, and visualization of spectra data. This paper proposes the use of ontologies to guide the composition of preprocessing and data mining tools and describes the approach through MS-Analyzer, a software tool for the integrated management, preprocessing and mining of spectra data on the Grid.

Sharing mass spectrometry data in a grid-based distributed proteomics laboratory

Data produced by mass spectrometry (MS) have been using in proteomics experiments to identify proteins or patterns in clinical samples that may be responsible for human diseases. MS-based proteomics is becoming a powerful, widely used technique to identify different molecular targets in different pathological contexts. Moreover, MS samples contain a huge amount of data; retrieving such information requires accessing to large volumes of data, thus an efficient organization is necessary both to reduce access time and to allow efficient knowledge extraction. Bioinformatics laboratories have been using more than one mass spectrometer to improve efficiency, largely increasing the volume of data obtained by experiments. Moreover, experimental data is enriched by observations and descriptions added by specialists through metadata. Thus, information retrieval of spectra data (and metadata describing them) inside a laboratory and among different laboratories requires large and scalable storage solutions, and high performance computational platforms. We present a software system for managing, sharing, and querying MS data in a distributed laboratory, using a spectra data management system, called SpecDB, where information retrieval is performed by using computational grid facilities. Information retrieval can be conducted either locally, by considering portions of spectra data, or in a distributed scenario, exploiting metadata and annotations about spectra datasets stored on the grid.

Preprocessing of mass spectrometry proteomics data on the grid

The combined use of mass spectrometry and data mining is a novel approach in proteomic pattern analysis for discovering novel biomarkers or identifying patterns and associations in proteomic profiles. Data produced by mass spectrometers are affected by errors and noise due to sample preparation and instrument approximation, so different preprocessing techniques need to be applied before analysis is conducted. We survey different techniques for spectra preprocessing, and we present a first design of a software tool that allows the preprocessing, management and analysis of mass spectrometry data on the Grid.

Improving protein secondary structure predictions by prediction fusion

Protein secondary structure prediction is still a challenging problem at today. Even if a number of prediction methods have been presented in the literature, the various prediction tools that are available on-line produce results whose quality is not always fully satisfactory. Therefore, a user has to know which predictor to use for a given protein to be analyzed. In this paper, we propose a server implementing a method to improve the accuracy in protein secondary structure prediction. The method is based on integrating the prediction results computed by some available on-line prediction tools to obtain a combined prediction of higher quality. Given an input protein p whose secondary structure has to be predicted, and a group of proteins F, whose secondary structures are known, the server currently works according to a two phase approach: (i) it selects a set of predictors good at predicting the secondary structure of proteins in F (and, therefore, supposedly, that of p as well), and (ii) it integrates the prediction results delivered for p by the selected team of prediction tools. Therefore, by exploiting our system, the user is relieved of the burden of selecting the most appropriate predictor for the given input protein being, at the same time, assumed that a prediction result at least as good as the best available one will be delivered. The correctness of the resulting prediction is measured referring to EVA accuracy parameters used in several editions of CASP.

A time series approach for clustering mass spectrometry data

Abstract Advanced statistical techniques and data mining methods have been recognized as a powerful support for mass spectrometry (MS) data analysis. Particularly, due to its unsupervised learning nature, data clustering methods have attracted increasing interest for exploring, identifying, and discriminating pathological cases from MS clinical samples. Supporting biomarker discovery in protein profiles has drawn special attention from biologists and clinicians. However, the huge amount of information contained in a single sample, that is, the high-dimensionality of MS data makes the effective identification of biomarkers a challenging problem. In this paper, we present a data mining approach for the analysis of MS data, in which the mining phase is developed as a task of clustering of MS data. Under the natural assumption of modeling MS data as time series, we propose a new representation model of MS data which allows for significantly reducing the high-dimensionality of such data, while preserving the relevant features. Besides the reduction of high-dimensionality (which typically affects effectiveness and efficiency of computational methods), the proposed representation model of MS data also alleviates the critical task of preprocessing the raw spectra in the whole process of MS data analysis. We evaluated our MS data clustering approach to publicly available proteomic datasets, and experimental results have shown the effectiveness of the proposed approach that can be used to aid clinicians in studying and formulating diagnosis of pathological states.

Automatic summarisation and annotation of microarray data

The study of biological processes within cells is based on the measurement of the activity of different molecules, in particular genes and proteins whose activities are strictly related. The activity of genes is measured through a systematic investigation carried out by microarrays. Such technology enables the investigation of all the genes of an organism in a single experiment, encoding meaningful biological information. Nevertheless, the preprocessing of raw microarray data needs automatic tools that standardise such phase in order to: (a) avoiding errors in analysis phases, and (b) making comparable the results of different laboratories. The preprocessing problem is as much relevant as considering results obtained from analysis platforms of different vendors. Nevertheless, there is currently a lack of tools that allow to manage and preprocess multivendor dataset. This paper presents a software platform (called GSAT, General-purpose Summarisation and Annotation Tool) able to manage and preprocess microarray data. The GSAT allows the summarisation, normalisation and annotation of multivendor microarray data, using web services technology. First experiments and results on Affymetrix data samples are also discussed. GSAT is available online at http://bioingegneria.unicz.it/m-cs as a standalone application or as a plugin of the TMEV microarray data analysis platform.

The EIPeptiDi tool: enhancing peptide discovery in ICAT-based LC MS/MS experiments

BackgroundIsotope-coded affinity tags (ICAT) is a method for quantitative proteomics based on differential isotopic labeling, sample digestion and mass spectrometry (MS). The method allows the identification and relative quantification of proteins present in two samples and consists of the following phases. First, cysteine residues are either labeled using the ICAT Light or ICAT Heavy reagent (having identical chemical properties but different masses). Then, after whole sample digestion, the labeled peptides are captured selectively using the biotin tag contained in both ICAT reagents. Finally, the simplified peptide mixture is analyzed by nanoscale liquid chromatography-tandem mass spectrometry (LC-MS/MS). Nevertheless, the ICAT LC-MS/MS method still suffers from insufficient sample-to-sample reproducibility on peptide identification. In particular, the number and the type of peptides identified in different experiments can vary considerably and, thus, the statistical (comparative) analysis of sample sets is very challenging. Low information overlap at the peptide and, consequently, at the protein level, is very detrimental in situations where the number of samples to be analyzed is high.ResultsWe designed a method for improving the data processing and peptide identification in sample sets subjected to ICAT labeling and LC-MS/MS analysis, based on cross validating MS/MS results. Such a method has been implemented in a tool, called EIPeptiDi, which boosts the ICAT data analysis software improving peptide identification throughout the input data set. Heavy/Light (H/L) pairs quantified but not identified by the MS/MS routine, are assigned to peptide sequences identified in other samples, by using similarity criteria based on chromatographic retention time and Heavy/Light mass attributes. EIPeptiDi significantly improves the number of identified peptides per sample, proving that the proposed method has a considerable impact on the protein identification process and, consequently, on the amount of potentially critical information in clinical studies. The EIPeptiDi tool is available at http://bioingegneria.unicz.it/~veltri/projects/eipeptidi/ with a demo data set.ConclusionEIPeptiDi significantly increases the number of peptides identified and quantified in analyzed samples, thus reducing the number of unassigned H/L pairs and allowing a better comparative analysis of sample data sets.

A framework for the atrial fibrillation prediction in electrophysiological studies

BACKGROUND AND OBJECTIVE Cardiac arrhythmias are disorders in terms of speed or rhythm in the hearts electrical system. Atrial fibrillation (AFib) is the most common sustained arrhythmia that affects a large number of persons. Electrophysiologic study (EPS) procedures are used to study fibrillation in patients; they consist of inducing a controlled fibrillation in surgical room to analyze electrical heart reactions or to decide for implanting medical devices (i.e., pacemaker). Nevertheless, the spontaneous induction may generate an undesired AFib, which may induce risk for patient and thus a critical issue for physicians. We study the unexpected AFib onset, aiming to identify signal patterns occurring in time interval preceding an event of spontaneous (i.e., not inducted) fibrillation. Profiling such signal patterns allowed to design and implement an AFib prediction algorithm able to early identify a spontaneous fibrillation. The objective is to increase the reliability of EPS procedures. METHODS We gathered data signals collected by a General Electric Healthcares CardioLab electrophysiology recording system (i.e., a polygraph). We extracted superficial and intracavitary cardiac signals regarding 50 different patients studied at the University Magna Graecia Cardiology Department. By studying waveform (i.e., amplitude and energy) of intracavitary signals before the onset of the arrhythmia, we were able to define patterns related to AFib onsets that are side effects of an inducted fibrillation. RESULTS A framework for atrial fibrillation prediction during electrophysiological studies has been developed. It includes a prediction algorithm to alert an upcoming AFib onset. Tests have been performed on an intracavitary cardiac signals data set, related to patients studied in electrophysiological room. Also, results have been validated by the clinicians, proving that the framework can be useful in case of integration with the polygraph, helping physicians in managing and controlling of patient status during EPS.

Toward an accurate prediction of inter-residue distances in proteins using 2D recursive neural networks

BackgroundProtein inter-residue contact maps provide a translation and rotation invariant topological representation of a protein. They can be used as an intermediary step in protein structure predictions. However, the prediction of contact maps represents an unbalanced problem as far fewer examples of contacts than non-contacts exist in a protein structure.In this study we explore the possibility of completely eliminating the unbalanced nature of the contact map prediction problem by predicting real-value distances between residues. Predicting full inter-residue distance maps and applying them in protein structure predictions has been relatively unexplored in the past.ResultsWe initially demonstrate that the use of native-like distance maps is able to reproduce 3D structures almost identical to the targets, giving an average RMSD of 0.5Å. In addition, the corrupted physical maps with an introduced random error of ±6Å are able to reconstruct the targets within an average RMSD of 2Å.After demonstrating the reconstruction potential of distance maps, we develop two classes of predictors using two-dimensional recursive neural networks: an ab initio predictor that relies only on the protein sequence and evolutionary information, and a template-based predictor in which additional structural homology information is provided. We find that the ab initio predictor is able to reproduce distances with an RMSD of 6Å, regardless of the evolutionary content provided. Furthermore, we show that the template-based predictor exploits both sequence and structure information even in cases of dubious homology and outperforms the best template hit with a clear margin of up to 3.7Å.Lastly, we demonstrate the ability of the two predictors to reconstruct the CASP9 targets shorter than 200 residues producing the results similar to the state of the machine learning art approach implemented in the Distill server.ConclusionsThe methodology presented here, if complemented by more complex reconstruction protocols, can represent a possible path to improve machine learning algorithms for 3D protein structure prediction. Moreover, it can be used as an intermediary step in protein structure predictions either on its own or complemented by NMR restraints.